Search

[Display omitted] • 1D zeolitic imidazolate framework-nanorod (ZIF-NR) has been prepared through facet engineering of the parental 2D ZIF-L. • The Co-NCR possesses rod-like morphology, metallic Co nanoparticle, high N content, and hierarchical porous structure. • Co-NCR exhibits superior ORR activity (E onset = 0.96 V, E 1/2 = 0.88 V). • Zinc-air battery assembled with Co-NCR exhibits high performance. The properties of metal–organic framework (MOF) nanocrystals are highly dependent on their sizes, morphologies, and exposed facets. Facet engineering of MOFs offers an efficient strategy to tailor the active sites and optimize the catalytic activity of both MOFs and their derivatives. In this study, we prepared 1D zeolitic imidazolate framework-nanorod (ZIF-NR) through facet engineering of the parental 2D ZIF-L. The introduction of cetyltrimethylammonium bromide (CTABr) surfactant into the synthesis solution hindered the crystal growth along the c -axis of leaf-like ZIF-L, resulting in the formation of 1D ZIF-NR. The derived Co nanoparticle encapsulated N doped carbon nanorod (denoted as Co-NCR) exhibited high activity and stability for electrocatalytic oxygen reduction reactions and Zn-air batteries. Facet engineering of a 2D MOF with a uniquely oriented layered structure demonstrates the possibility of designing novel electrocatalysts. [ABSTRACT FROM AUTHOR]

Ce-promoted NiMgAl mixed-oxide (NiCex-C, x = 0, 1, 5, 10) catalysts were prepared from the quaternary hydrotalcite precursors for CO2 hydrogenation to methane. By engineering the Ce contents, NiCe5-C showed its prior catalytic performance in low-temperature CO2 hydrogenation, being about three times higher than that of the Ce-free NiCe0-C catalyst (turnover frequency of NiCe5-C and NiCe0-C: 11.9 h−1 vs. 3.9 h−1 @ 225 °C). With extensive characterization, it was found that Ce dopants promoted the reduction of NiO by adjusting the interaction between Ni and Mg(Ce)AlOx support. The highest ratio of surface Ni0/(Ni2+ + Ni0) was obtained over NiCe5-C. Meanwhile, the surface basicity was tailored with Ce dopants. The strongest medium-strength basicity and highest capacity of CO2 adsorption was achieved on NiCe5-C with 5 wt.% Ce content. The TOF tests indicated a good correlation with medium-strength basicity over the NiCex-C samples. The results showed that the high medium-strength and Ce-promoted surface Ni0 species endows the enhanced low-temperature catalytic performance in CO2 hydrogenation to methane. [ABSTRACT FROM AUTHOR]

Objective: Everyone in life will experience resource scarcity, which causes self-discrepancy. It is widely known that individuals participate in reactive consumption to solve the problems of self-discrepancy and resources scarcity. This kind of consumption may be symbolically related to the essence of the resource scarcity or may occur in an unrelated domain. This study proposes a theory for "filling up" one's resource scarcity through high-intensity sensory consumption (HISC). Methods: We used different methods, including one-way analysis of variance (ANOVA), linear regression, mediating effect, and moderating effect, to test the four hypotheses. Four experiments in the study were conducted from May 2022 and August 2022 and involved undergraduates from a university and volunteers recruited online. All participants are adults and verbally agree to participate voluntarily. Study 1a (N = 96 (male 47, female 49), participants from a business school in China) measured resource scarcity in the laboratory experiments and verified the effect of resource scarcity on consumer HISC preference by using linear regression (H1). Study 1b (N = 191 (male 98, female 93), students and teachers from a university in China) measured resource scarcity in the laboratory experiments and manipulated positively and negatively valenced experiences. Using the PROCESS SPSS Mode l, we verified that negatively valenced stimuli also lead to higher levels of arousal, which in turn restores the self-discrepancy caused by resource scarcity (H2). Study 2 (an online experiment, N = 182 (male 91, female 91), participants from China) manipulated the resource scarcity in a color sensory stimulant context, replicating the preliminary effect and examined the mediating effect of the self-worth by using the PROCESS SPSS Mode 4 (H3). Study 3 (an online experiment, N = 251 (male 125, female 126), participants from China) manipulated resource scarcity and self-acceptance in the tactile sensory experience, and tested the moderating effect of self-acceptance by using the PROCESS SPSS Mode 8 (H4). Results: Four studies suggest that not only do individuals facing resources scarcity prefer HISC but also that this consumption is mediated and moderated by self-worth and self-acceptance, respectively. This preference for HISC is negated when individuals have high self-acceptance traits. The findings are tested in the auditory domain (as evidenced by a propensity for louder volume), the visual domain (as evidenced by a propensity for more intense colors), and the tactile domain (as evidenced by a propensity for more intense need for touch). The findings also demonstrate that individual preferences for HISC is shown to operate regardless of the valence (positive valence vs. negative valence) of the sensory consumption. Conclusions: Across four experiments, we find that individuals who are subjected to resource scarcity show a preference for high-intensity sensory consumption in the auditory, visual, and tactile domains. We also find that both negatively and positively valenced sensory stimuli have the same impact on resource-scarce individuals' preference for HISC. Furthermore, we demonstrate that the sense of self-worth significantly mediates the effect of resource scarcity on HISC. Finally, we reveal that self-acceptance moderates the effect of resource scarcity on HISC preference. [ABSTRACT FROM AUTHOR]

A novel Gram-stain-positive, aerobic bacterial strain, designated AK-R2A1-2 T, was isolated from the surface-sterilized needle leaves of an Abies koreana tree. Strain AK-R2A1-2 T had 97.3% and 96.7% 16S rRNA gene sequence similarities with Subtercola boreus K300T and Subtercola lobariae 9583bT, respectively, but formed a distinct phyletic lineage from these two strains. Growth of strain AK-R2A1-2 T was observed at 4–25 °C at pH 5.0–8.0. Strain AK-R2A1-2 T contained menaquinone 9 (MK-9) and menaquinone 10 (MK-10) as the predominant respiratory quinones. The major cellular fatty acids were anteiso-C15:0 and summed feature 8 (C18:1ω7c or/and C18:1ω6c), and the polar lipids included diphosphatidylglycerol (DPG) and three unknown aminolipids, AKL2, AKL3, and AKL4. The complete genome of strain AK-R2A1-2 T was sequenced to understand the genetic basis of its survival at low temperatures. Multiple copies of cold-associated genes involved in cold-active chaperon, stress response, and DNA repair supported survival of the strain at low temperatures. Strain AK-R2A1-2 T was also able to significantly improve rice seedling growth under low temperatures. Thus, this strain represents a novel species of the genus Subtercola, and the proposed name is Subtercola endophyticus sp. nov. The type strain is AK-R2A1-2 T (= KCTC 49721 T = GDMCC 1.2921 T). [ABSTRACT FROM AUTHOR]

Cross-media retrieval (CMR) offers a flexible retrieval experience across multiple modalities. Existing CMR approaches are constrained by the assumption that the paired modalities are available in training, and they leverage the data of all modalities to obtain a common representation. However, as dealing with the data from new modality, the previous all modalities need to be re-trained, compromising the flexibility and practicality of CMR. In this paper, we propose an approach termed learning Modality-Agnostic Representation for Scalable cross-media retrieval (MARS), which allows each modality to be trained independently. To be specific, MARS treats the label information as a distinct modality, and introduces a label parsing module LabNet to generate semantic representation for correlating different modalities. Meanwhile, MARS constructs the modality-specific representation module DataNet to obtain the modality-shared representation and modality-exclusive representation equipped with unbiased semantic classification. Technically, for the first modality, we jointly train the LabNet and its DataNet to preserve the semantic similarity between the Label-derived representation and the modality-shared representation. For new modalities, MARS employs the well-learned LabNet to extract the representation in labels, and then such representation is served as the privilege to guide the associated DataNet training via the same objective. Furthermore, we assign the same classifier to the representation module of all modalities for better semantic alignment. With the above schema, the obtained modality-shared representation is considered to be modality-agnostic. Extensive experiments on several benchmark multi-modality datasets demonstrate that the proposed MARS achieves better results than existing methods. [ABSTRACT FROM AUTHOR]

Domain Incremental Learning (DIL) focuses on handling complex domain shifts of a continuous data stream for visual tasks such as image classification and image segmentation. In real life, severe domain gaps in DIL are generated from various sources such as data style shifts, data quality degradation, environment changes, and so on. The well-known catastrophic forgetting issue in DIL becomes even more critical when simultaneously considering multiple sources of domain shifts. In this paper, we propose a unified and effective paradigm named Compositional Prompting (C-Prompt) to mitigate the critical forgetting challenge in DIL for image classification tasks. Unlike a popular type of conventional DIL approaches that need to retain abundant exemplars from the old domains, our exemplar-free C-Prompt leverages a prompt-guided Batch-wise Exponential Moving Average (BEMA) strategy to adaptively consolidate learned knowledge without retaining any exemplars. A set of prompts shared across different domains is designed to estimate the knowledge shifts for automatically balancing knowledge acquisition and forgetting. To enhance the learning ability, our proposed C-Prompt explores a domain-specific pool of learnable prompts for each domain, and all the prompt pools are further exploited in a cross-domain compositional manner to facilitate inference. Since the latest prompting-based DIL methods aim to learn one individual prompt for each domain, they always suffer from critical performance degradation caused by the incorrect prediction of domain index during inference and the limited learning capacity by using a single prompt per domain. Instead, our C-Prompt can not only readily acquire domain-specific knowledge but also exploit domain-shared knowledge. Extensive experiments on various large-scale multi-domain benchmarks have demonstrated the superiority of our proposed C-Prompt compared with state-of-the-art methods. Code is available at https://github.com/zhoujiahuan1991/IJCV2024-C-Prompt. [ABSTRACT FROM AUTHOR]

A bacterial strain, designated J12C1-MA-4T, was isolated from liquid culture of the dinoflagellate Ceratoperidinium margalefii. The bacterium was Gram-negative, aerobic, and rod-shaped. Oxidase and catalase were positive. Optimal growth was observed at 30 °C, pH 7.0, in the presence of 1% (w/v) NaCl. Phylogenetic analyses based on 16S rRNA gene and a 92 core gene set suggested that the strain J12C1-MA-4T belongs to the family Rhodobacteraceae in the class Alphaproteobacteria and represents a taxon separated from other genera. 16S rRNA gene sequence of the strain J12C1-MA-4T showed high similarities to Loktanella ponticola KCTC 42133T (95.7%), Pseudooctadecabacter jejudonensis KCTC 32525T (95.5%) and Jannaschia helgolandensis KCTC 12191T (95.3%). The genome length of strain J12C1-MA-4T was 3,621,968 bp with a DNA G + C content of 64.48 mol%. The major cellular fatty acids of strain J12C1-MA-4T were summed feature 8 (comprising C18:1ω7c and/or C18:1ω6c) (> 10%). Phosphatidylglycerol (PG), phosphatidylcholine (PC), phospholipids (PL), lipids 1 (L1) and aminolipid (AL) were shown to be the major polar lipids. The sole predominant isoprenoid quinone was Q-10. Based on phylogenetic, phenotypic, chemotaxonomic and genomic features, we propose that strain J12C1-MA-4T represent a novel species in the novel genus of the family Rhodobacteraceae, with the proposed name Gymnodinialimonas ceratoperidinii gen. nov., sp. nov.. The type strain is J12C1-MA-4T (=KCTC 82770T =GDMCC 1.2729T). [ABSTRACT FROM AUTHOR]

Salinization of land is globally increasing due to climate change, and salinity stress is an important abiotic stressor that adversely affects agricultural productivity. In this study, we assessed a halotolerant endophytic bacterium, Pseudoxanthomonas sp. JBR18, for its potential as a plant growth-promoting agent with multiple beneficial properties. The strain exhibited tolerance to sodium chloride concentration of up to 7.5 % in the R2A medium. In vitro evaluation revealed that strain JBR18 possessed proteolytic, protease (EC 3.4), and cellulase (EC 3.2.1.4) activities, as well as the ability to produce indole-acetic acid, proline, and exopolysaccharides. Compared with the controls, co-cultivation of Arabidopsis seedlings with the strain JBR18 improved plant growth, rosette size, shoot and root fresh weight, and chlorophyll content under salinity stress. Moreover, JBR18-inoculated seedlings showed lower levels of malondialdehyde, reactive oxygen species, and Na+ uptake into plant cells under salt stress but higher levels of K+. Additionally, seedlings inoculated with JBR18 exhibited a delayed response time and quantity of salt-responsive genes RD29A , RD29B , RD20 , RD22 , and KIN1 under salt stress. These multiple effects suggest that Pseudoxanthomonas sp. JBR18 is a promising candidate for mitigating the negative impacts of salinity stress on plant growth. Our findings may assist in future efforts to develop eco-friendly strategies for managing abiotic stress and enhancing plant tolerance to salt stress. [Display omitted] • JBR18 significantly increased Arabidopsis biomass under salt and non-salt stress. • JBR18 inoculation stimulated indole-3-acetic acid production in Arabidopsis roots. • JBR18 decreased ROS leves and Na + ion uptake into plant cells under salt stress. • JBR18 reduced the salt stress response gene levels in Arabidopsis under salt stress. • JBR18 is a promising microbial agent that can mitigate the negative effects of salinity stress on plant growth. [ABSTRACT FROM AUTHOR]

Correlation learning among different types of multimedia data, such as visual and textual content, faces huge challenges from two important perspectives, namely, cross modal and cross domain. Cross modal means the heterogeneous properties of different types of multimedia data, where the data from different modalities have inconsistent distributions and representations. This situation leads to the first challenge: cross-modal similarity measurement. Cross domain means the multisource property of multimedia data from various domains, in which data from new domains arrive continually, leading to the second challenge: model storage and retraining. Therefore, correlation learning requires a cross-modal continual learning approach, in which only the data from the new domains are used for training, but the previously learned correlation capabilities are preserved. To address the above issues, we introduce the idea of life-long learning into visual-textual cross-modal correlation modeling and propose a visual-textual life-long knowledge distillation (VLKD) approach. In this study, we construct a hierarchical recurrent network that can leverage knowledge from both semantic and attention levels through adaptive network expansion to support cross-modal retrieval in life-long scenarios across various domains. The results of extensive experiments performed on multiple cross-modal datasets with different domains verify the effectiveness of the proposed VLKD approach for life-long cross-modal retrieval. [ABSTRACT FROM AUTHOR]

When glancing at an image, human can infer what is hidden in the image beyond what is visually obvious, such as objects’ functions, people’s intents and mental states. However, such a visual reasoning paradigm is tremendously difficult for computer, requiring knowledge about how the world works. To address this issue, we propose Commonsense Knowledge based Reasoning Model (CKRM) to acquire external knowledge to support Visual Commonsense Reasoning (VCR) task, where the computer is expected to answer challenging visual questions. Our key ideas are: (1) To bridge the gap between recognition-level and cognition-level image understanding, we inject external commonsense knowledge via multi-level knowledge transfer network, achieving cell-level, layer-level and attention-level joint information transfer. It can effectively capture knowledge from different perspectives, and perceive common sense of human in advance. (2) To further promote image understanding at cognitive level, we propose a knowledge based reasoning approach, which can relate the transferred knowledge to visual content and compose the reasoning cues to derive the final answer. Experiments are conducted on the challenging visual commonsense reasoning dataset VCR to verify the effectiveness of our proposed CKRM approach, which can significantly improve reasoning performance and achieve the state-of-the-art accuracy. [ABSTRACT FROM AUTHOR]

Existing cross-media retrieval methods are usually conducted under the supervised setting, which need lots of annotated training data. Generally, it is extremely labor-consuming to annotate cross-media data. So unsupervised cross-media retrieval is highly demanded, which is very challenging because it has to handle heterogeneous distributions across different media types without any annotated information. To address the above challenge, this paper proposes Domain Adaptation with Scene Graph (DASG) approach, which transfers knowledge from the source domain to improve cross-media retrieval in the target domain. Our DASG approach takes Visual Genome as the source domain, which contains image knowledge in the form of scene graph. The main contributions of this paper are as follows: First, we propose to address unsupervised cross-media retrieval by domain adaptation. Instead of using the labor-consuming annotated information of cross-media data in the training stage, our DASG approach learns cross-media correlation knowledge from Visual Genome, and then transfers the knowledge to cross-media retrieval through media alignment and distribution alignment. Second, our DASG approach utilizes fine-grained information via scene graph representation to enhance generalization capability across domains. The generated scene graph representation builds (subject $\rightarrow $ relationship $\rightarrow $ object) triplets by exploiting objects and relationships within image and text, which makes the cross-media correlation more precise and promotes unsupervised cross-media retrieval. Third, we exploit the related tasks including object and relationship detection for learning more discriminative features across domains. Leveraging the semantic information of objects and relationships improves cross-media correlation learning for retrieval. Experiments on two widely-used cross-media retrieval datasets, namely Flickr-30K and MS-COCO, show the effectiveness of our DASG approach. [ABSTRACT FROM AUTHOR]

Text-to-image synthesis is to generate images with the consistent content as the given text description, which is a highly challenging task with two main issues: visual reality and content consistency. Recently, it is available to generate images with high visual reality due to the significant progress of generative adversarial networks. However, translating text description to image with high content consistency is still ambitious. For addressing the above issues, it is reasonable to establish a transitional space with interpretable representation as a bridge to associate text and image. So we propose a text-to-image synthesis approach named Bridge-like Generative Adversarial Networks (Bridge-GAN). Its main contributions are: (1) A transitional space is established as a bridge for improving content consistency, where the interpretable representation can be learned by guaranteeing the key visual information from given text descriptions. (2) A ternary mutual information objective is designed for optimizing the transitional space and enhancing both the visual reality and content consistency. It is proposed under the goal to disentangle the latent factors conditioned on text description for further interpretable representation learning. Comprehensive experiments on two widely-used datasets verify the effectiveness of our Bridge-GAN with the best performance. [ABSTRACT FROM AUTHOR]

Multi-media data including image, video, text, audio, and 3D model, has been fast emerging on the Internet. Jointly correlating the data of various media types is a challenging task. With the considerable learning ability of deep network, existing works mainly construct multi-pathway network to learn cross-media correlation, where each pathway is for one media type. However, with number of media types increasing, existing methods face the problems of high repetition and complexity, leading to overfitting and poor generalization ability, which makes adverse effect on correlation learning. For addressing the above issues, we propose cross-media deep compression and regularization (CDCR) approach for quintuple-media joint correlation learning: 1) cross-media partial weight-sharing networks is proposed, where a part of parameters are commonly shared among multiple pathways, to exploit common characteristics across different media types for capturing intrinsic cross-media correlation; 2) we propose media-adaptive network pruning to drop connections between weakly-correlated neurons, which can emphasize media-specific characteristics adaptively; and 3) cross-media network regularization is proposed to utilize relationships among quintuple-media data, which can guarantee generalization ability and enhance intra-media and inter-media correlation. The experiments verify the effectiveness of our approach, which outperforms the state-of-the-art methods on two very challenging datasets, including a large-scale dataset PKU XMediaNet with more than 100 000 quintuple-media instances. [ABSTRACT FROM AUTHOR]

In a badminton game, dexterous fine hand movements are essential for the player to perform with extraordinary skill. Thus, this paper proposes a smart glove with proper sensing functions for finger pressures and bending angles for this sport to measure the kinematics and kinetics during the forehand grip and backhand grip. Twenty-one pressure sensors and eleven flex sensors are employed to construct the smart glove, and the measured parameters are then visualized to analyze different grip movements. An athlete and an amateur were selected for the gripping experiments and the measurements was recorded by the smart glove. During forehand gripping, although the total power generated by the amateur (33 N) was higher than for the athlete (29 N), the final shot strength was still unsatisfactory due to the disorderly arranged finger flex angle. In backhand gripping, the incorrectly laid index finger reduced the thumb power of the amateur to 3 N, which weakened the control ability and led to a distorted flying trajectory of the shuttlecock. The differences in kinematics and kinetics between the athlete and the amateur can provide a reference for the badminton learners to perform the correct technical movements. By use of this approach, the proposed system may grant trainers and trainees a better approach to analyze the grip movement. [ABSTRACT FROM AUTHOR]

Video captioning is a significant challenging task in computer vision and natural language processing, aiming to automatically describe video content by natural language sentences. Comprehensive understanding of video is the key for accurate video captioning, which needs to not only capture the global content and salient objects in video, but also understand the spatio-temporal relations of objects, including their temporal trajectories and spatial relationships. Thus, it is important for video captioning to capture the objects’ relationships both within and across frames. Therefore, in this paper, we propose an object-aware spatio-temporal graph (OSTG) approach for video captioning. It constructs spatio-temporal graphs to depict objects with their relations, where the temporal graphs represent objects’ inter-frame dynamics, and the spatial graphs represent objects’ intra-frame interactive relationships. The main novelties and advantages are: (1) Bidirectional temporal alignment: Bidirectional temporal graph is constructed along and reversely along the temporal order to perform bidirectional temporal alignment for objects across different frames, which provides complementary clues to capture the inter-frame temporal trajectories for each salient object. (2) Graph based spatial relation learning: Spatial relation graph is constructed among objects in each frame by considering their relative spatial locations and semantic correlations, which is exploited to learn relation features that encode intra-frame relationships for salient objects. (3) Object-aware feature aggregation: Trainable VLAD (vector of locally aggregated descriptors) models are deployed to perform object-aware feature aggregation on objects’ local features, which learn discriminative aggregated representations for better video captioning. A hierarchical attention mechanism is also developed to distinguish contributions of different object instances. Experiments on two widely-used datasets, MSR-VTT and MSVD, demonstrate our proposed approach achieves state-of-the-art performances in terms of BLEU@4, METEOR and CIDEr metrics. [ABSTRACT FROM AUTHOR]

The rapidly developing information technology leads to a fast growth of visual and textual contents, and it comes with huge challenges to make correlation and perform cross-media retrieval between images and sentences. Existing methods mainly explore cross-media correlation from either global-level instances as the whole images and sentences, or local-level fine-grained patches as the discriminative image regions and key words, which ignore the complementary information from the relation between local-level fine-grained patches. Naturally, relation understanding is highly important for learning cross-media correlation. People focus on not only the alignment between discriminative image regions and key words, but also their relations lying in the visual and textual context. Therefore, in this paper, we propose Multi-level Adaptive Visual-textual Alignment (MAVA) approach with the following contributions. First, we propose cross-media multi-pathway fine-grained network to extract not only the local fine-grained patches as discriminative image regions and key words, but also visual relations between image regions as well as textual relations from the context of sentences, which contain complementary information to exploit fine-grained characteristics within different media types. Second, we propose visual-textual bi-attention mechanism to distinguish the fine-grained information with different saliency from both local and relation levels, which can provide more discriminative hints for correlation learning. Third, we propose cross-media multi-level adaptive alignment to explore global, local and relation alignments. An adaptive alignment strategy is further proposed to enhance the matched pairs of different media types, and discard those misalignments adaptively to learn more precise cross-media correlation. Extensive experiments are conducted to perform image-sentence matching on 2 widely-used cross-media datasets, namely Flickr-30K and MS-COCO, comparing with 10 state-of-the-art methods, which can fully verify the effectiveness of our proposed MAVA approach. [ABSTRACT FROM AUTHOR]

Fine-grained image classification aims to distinguish subcategories belonging to the same basic-level category, such as 200 subcategories belonging to bird. It is a challenging problem in computer vision and multimedia field due to: attribute similarity (e.g. color and texture) among different subcategories and attribute variance (e.g. pose and viewpoint) in the same subcategory. Attribute similarity causes the difficulty to classify different subcategories even for human, while attribute variance causes the learned feature representations chaotic and confused. Naturally, classification can benefit from a hierarchy of subcategories: since going to a coarser granularity leverages high-level semantic features, while going to a finer granularity leverages discriminative and subtle features. Therefore, we propose an attribute hierarchy based multi-task learning (AHMTL) approach, and its main novelties are: (1) Attribute hierarchy: We reassign all images to multi-granularity subcategories automatically, namely coarse-grained, fine-grained and ultra-fine-grained subcategories. Similar fine-grained subcategories are reassigned to the same coarse-grained subcategory according to their attribute similarity, which pays more attention to the high-level semantic feature representations. Simultaneously, the same fine-grained subcategory but with different attributes are divided into different ultra-fine-grained subcategories, which can obtain more discriminative and subtle feature representations for attribute variance. (2) Multi-task learning: A multi-task learning framework is designed to effectively learn robust feature representations by jointly optimizing coarse-grained, fine-grained and ultra-fine-grained image classification tasks. These three level tasks learn coarse to fine feature representations, meaning high-level semantic to subtle features, which can regularize and boost each other to prevent overfitting, and the mutual promotion of them ensures feature representations more discriminative. Compared with more than 10 state-of-the-art methods on two widely-used CUB-200-2011 and Cars-196 datasets, our AHMTL approach achieves the best classification performance. [ABSTRACT FROM AUTHOR]

The heterogeneity gap leads to inconsistent distributions and representations between image and text, which rises a challenging task to measure their similarities and construct cross-media correlation between them. The existing works mainly model the cross-media correlation in a common subspace, which causes insufficient correlation modeling in such third-party subspace with intermediate unidirectional transformation. Inspired by the recent advances of neural machine translation, which aims to establish a corresponding relationship between two entirely different languages, we can naturally discover that it has striking common characteristic with cross-media correlation learning to consider image and text as bilingual pairs, where the image is treated as a special kind of language to provide visual description, so that bidirectional transformation can be conducted between image and text to effectively explore cross-media correlation in the feature space of each media type. Thus, we propose a reinforced cross-media bidirectional translation (RCBT) approach to model the correlation between visual and textual descriptions. First, cross-media bidirectional translation mechanism is proposed to conduct direct transformation between the bilingual pairs of visual and textual descriptions bidirectionally, where the cross-media correlation can be effectively captured in both feature spaces of image and text through bidirectional translation training. Second, cross-media context-aware network with residual attention is proposed to exploit the rich spatial and temporal context hints with cross-media convolutional recurrent neural network, which can lead to more precise correlation learning for promoting bidirectional translation process. Third, cross-media reinforcement learning is proposed to perform a two-agent communication game played as a round between image and text to boost the bidirectional translation process, and we further extract inter-media and intra-media reward signals to provide complementary clues for learning cross-media correlation. Extensive experiments are conducted on cross-media retrieval to verify the effectiveness of our proposed RCBT approach, compared with 11 state-of-the-art methods on three cross-media datasets. [ABSTRACT FROM AUTHOR]

Existing cross-media retrieval methods are mainly based on the condition where the training set covers all the categories in the testing set, which lack extensibility to retrieve data of new categories. Thus, zero-shot cross-media retrieval has been a promising direction in practical application, aiming to retrieve data of new categories (unseen categories), only with data of limited known categories (seen categories) for training. It is challenging for not only the heterogeneous distributions across different media types, but also the inconsistent semantics across seen and unseen categories need to be handled. To address the above issues, we propose dual adversarial distribution network (DADN), to learn common embeddings and explore the knowledge from word-embeddings of different categories. The main contributions are as follows. First, zero-shot cross-media dual generative adversarial networks architecture is proposed, in which two kinds of generative adversarial networks (GANs) for common embedding generation and representation reconstruction form dual processes. The dual GANs mutually promote to model semantic and underlying structure information, which generalizes across different categories on heterogeneous distributions and boosts correlation learning. Second, distribution matching with maximum mean discrepancy criterion is proposed to combine with dual GANs, which enhances distribution matching between common embeddings and category word-embeddings. Finally, adversarial inter-media metric constraint is proposed with an inter-media loss and a quadruplet loss, which further model the inter-media correlation information and improve semantic ranking ability. The experiments on four widely used cross-media datasets demonstrate the effectiveness of our DADN approach. [ABSTRACT FROM AUTHOR]

Fine-grained visual categorization is to recognize hundreds of subcategories belonging to the same basic-level category, which is a highly challenging task due to the quite subtle and local visual distinctions among similar subcategories. Most existing methods generally learn part detectors to discover discriminative regions for better categorization performance. However, not all parts are beneficial and indispensable for visual categorization, and the setting of part detector number heavily relies on prior knowledge as well as experimental validation. As is known to all, when we describe the object of an image via textual descriptions, we mainly focus on the pivotal characteristics and rarely pay attention to common characteristics as well as the background areas. This is an involuntary transfer from human visual attention to textual attention, which leads to the fact that textual attention tells us how many and which parts are discriminative and significant to categorization. So, textual attention could help us to discover visual attention in the image. Inspired by this, we propose a fine-grained visual-textual representation learning (VTRL) approach, and its main contributions are: 1) fine-grained visual-textual pattern mining devotes to discovering discriminative visual-textual pairwise information for boosting categorization performance through jointly modeling vision and text with generative adversarial networks, which automatically and adaptively discovers discriminative parts and 2) VTRL jointly combines visual and textual information, which preserves the intra-modality and inter-modality information to generate complementary fine-grained representation, as well as further improves categorization performance. Comprehensive experimental results on the widely used CUB-200-2011 and Oxford Flowers-102 datasets demonstrate the effectiveness of our VTRL approach, which achieves the best categorization accuracy compared with the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Fine-grained visual categorization (FGVC) aims to discriminate similar subcategories that belong to the same superclass. Since the distinctions among similar subcategories are quite subtle and local, it is highly challenging to distinguish them from each other even for humans. So the localization of distinctions is essential for fine-grained visual categorization, and there are two pivotal problems: (1) Which regions are discriminative and representative to distinguish from other subcategories? (2) How many discriminative regions are necessary to achieve the best categorization performance? It is still difficult to address these two problems adaptively and intelligently. Artificial prior and experimental validation are widely used in existing mainstream methods to discover which and how many regions to gaze. However, their applications extremely restrict the usability and scalability of the methods. To address the above two problems, this paper proposes a multi-scale and multi-granularity deep reinforcement learning approach (M2DRL), which learns multi-granularity discriminative region attention and multi-scale region-based feature representation. Its main contributions are as follows: (1) Multi-granularity discriminative localization is proposed to localize the distinctions via a two-stage deep reinforcement learning approach, which discovers the discriminative regions with multiple granularities in a hierarchical manner ("which problem"), and determines the number of discriminative regions in an automatic and adaptive manner ("how many problem"). (2) Multi-scale representation learning helps to localize regions in different scales as well as encode images in different scales, boosting the fine-grained visual categorization performance. (3) Semantic reward function is proposed to drive M2DRL to fully capture the salient and conceptual visual information, via jointly considering attention and category information in the reward function. It allows the deep reinforcement learning to localize the distinctions in a weakly supervised manner or even an unsupervised manner. (4) Unsupervised discriminative localization is further explored to avoid the heavy labor consumption of annotating, and extremely strengthen the usability and scalability of our M2DRL approach. Compared with state-of-the-art methods on two widely-used fine-grained visual categorization datasets, our M2DRL approach achieves the best categorization accuracy. [ABSTRACT FROM AUTHOR]

There is a growing demand for positioning actuators with a higher resolution, accuracy, speed, and driving force. Various piezoelectric actuators have been proposed to meet these requirements; however, they all have inherent limitations. This paper presents a novel high-performance piezoelectric actuator that can overcome the limitations of existing designs. It is based on the stick–slip actuation principle and makes use of coupling motions of the proposed triangular driving mechanism to generate a clamping action during the “stick” phase and a releasing action during the “slip” phase. Unlike existing driving mechanisms based on similar principles, the proposed actuator can employ its unique structure to amplify the clamping force and the related driving force by using a large design triangular angle. Apart from its superior performance in driving force, it is interestingly found that its driving speed performance also improves when the design angle is increased. Finite-element analysis and experiments are carried out to justify the superior performance of the proposed actuator. In comparison with existing actuator prototypes based on similar principles, a prototype of the proposed actuator, even driven with a lower input voltage, achieves an 11 times larger driving load and a 3 times higher free-load driving speed. [ABSTRACT FROM AUTHOR]

Exploring electrocatalysts with high-efficiency oxygen reduction reaction (ORR) is significant for practical applications of fuel cells and metal–air batteries. In this work, a twisted core@shell material has been prepared with helical polypyrrole nanotubes (HPPys) as the core and coordination polymers (CPs) as the shell. After the pyrolysis process, a dense twisted carbon layer was formed by the reaction of CP and HPPy at its interface under Ar. The derived twisted carbonaceous nanotube exhibits good performance in both electrocatalytic ORR and OER. When used as the air-electrode in a flexible Zn–air battery, the battery shows good performance and stability. [ABSTRACT FROM AUTHOR]

Hashing methods have been widely used for efficient similarity retrieval on large scale image database. Traditional hashing methods learn hash functions to generate binary codes from hand-crafted features, which achieve limited accuracy since the hand-crafted features cannot optimally represent the image content and preserve the semantic similarity. Recently, several deep hashing methods have shown better performance because the deep architectures generate more discriminative feature representations. However, these deep hashing methods are mainly designed for supervised scenarios, which only exploit the semantic similarity information, but ignore the underlying data structures. In this paper, we propose the semi-supervised deep hashing approach, to perform more effective hash function learning by simultaneously preserving semantic similarity and underlying data structures. The main contributions are as follows: 1) We propose a semi-supervised loss to jointly minimize the empirical error on labeled data, as well as the embedding error on both labeled and unlabeled data, which can preserve the semantic similarity and capture the meaningful neighbors on the underlying data structures for effective hashing. 2) A semi-supervised deep hashing network is designed to extensively exploit both labeled and unlabeled data, in which we propose an online graph construction method to benefit from the evolving deep features during training to better capture semantic neighbors. To the best of our knowledge, the proposed deep network is the first deep hashing method that can perform hash code learning and feature learning simultaneously in a semi-supervised fashion. Experimental results on five widely-used data sets show that our proposed approach outperforms the state-of-the-art hashing methods. [ABSTRACT FROM AUTHOR]

The current study examined the factor structure and construct validity of lexical complexity and also unpacked its role in predicting perceived writing quality based on one data set. This study sampled 220 argumentative writings scored by expert raters from the International Corpus Network of Asian Learners of English (ICNALE) and used Coh-Metrix to compute the indices that tap into multiple constructs of lexical complexity. The results of Confirmatory Factor Analysis (CFA) confirmed the hypothesized factor structure and construct validity of lexical sophistication and lexical diversity. The results of Structural Equation Modeling (SEM) demonstrated that lexical sophistication and lexical diversity measured by computational indices significantly predicted human assessment of L2 writing quality at two language proficiency levels. On the other hand, human judgment on L2 writing quality was found more concerned with more perceivable text features such as linguistic errors and text length than the fine-grained features that tap into lexical complexity. Interpretations as to the small albeit significant variance explained by lexical complexity are provided in the discussion section. This study provides empirical evidence for understanding human scoring processes and implications for fine-tuning AES models to enhance their interpretability. • Factor structure and construct validity of multiple constructs tapping into lexical complexity. • Significant albeit small predictive power of lexical sophistication and lexical diversity in assessing writing. • Implications for fine-tuning AES systems to enhance interpretability. [ABSTRACT FROM AUTHOR]

Cognate awareness is the ability to recognise cognate relationships in languages, which requires activation of vocabulary knowledge in both languages. Different types of cognates may have various degrees of contributions to vocabulary knowledge and reading comprehension. Few studies have explored the role of different cognate types in vocabulary knowledge and reading across languages, especially in the L3 learner population. To this end, the present study examined the role of different facets of cognate awareness in L3 vocabulary and reading acquisition. Chinese college students of L2 English and L3 French (N = 121) participated in the study, all of whom had learned L2 English for over nine years and L3 French for at least one semester in classrooms. They were required to complete a researcher-developed English-French cognate test, a French vocabulary test and a French reading test. The English-French cognate test contained three types of cognates (same, similar, and false types). The French vocabulary test consisted of a vocabulary size subtest and a Word Associates subtest to measure breadth and depth knowledge. Results of path analysis revealed indirect effects of two types of cognate awareness (similar and false types) on L3 French reading comprehension via the mediation of vocabulary knowledge. [ABSTRACT FROM AUTHOR]

In this paper, a smooth impact drive mechanism (SIDM) actuation method for flapping wing mechanism which can be used in bio-inspired micro air vehicles is proposed. The SIDM-actuated flapping wing mechanism is composed of a base, a linear actuator, and a dual slider-crank mechanism. Based on the driving principle of SIDM, the linear actuator can generate a reciprocating linear motion with a long travel range. Via the dual slider-crank mechanism, the reciprocating linear motion can be converted to a flapping motion with a relatively high frequency. In addition, due to no gearbox used in the design, the SIDM-actuated flapping wing mechanism can be made with small volume and light weight. [ABSTRACT FROM AUTHOR]

A novel κ -carrageenase gene ( Cly-κ-car ) was cloned and heterologously expressed from marine bacterium Cellulophaga lytica strain N5-2. The gene comprised an open reading frame of 1488 bp encoding 495 amino acid residues. The deduced protein had a calculated molecular weight of 55.24 kDa with an estimated isoelectric point of 9.90. Multiple alignment analysis revealed that Cly-κ- CAR shared identity with κ -carrageenases from Zobellia sp. M-2 (46%), Zobellia galactanivorans (42%) and Rhodopirellula islandica (38%). Recombinant Cly - κ -CAR (R- Cly - κ -CAR) had maximum specific activity of 620.08 U/mg at 35 °C, pH 7.0, 0.7% κ -carrageenan and in the presence of 0.6% NaCl. It retained >75% of its initial activity after heat treatment below 35 °C for 2 h. More than 50% of its activity was maintained after incubation at pH 5.0-8.0 and 4 °C for 6 h. The K m and V max values for κ -carrageenan were 0.94 mg/ml and 13.42 mM/min/mg, respectively. Thin layer chromatographic analysis of the R- Cly - κ -CAR hydrolysis products revealed that the enzyme hydrolyzed κ- carrageenan into neo- κ -carraoctaose and neo- κ -carrahexaose. R- Cly - κ -CAR is a novel κ -carrageenase enzyme and could be a valuable tool to produce high degree of polymerization κ -carrageenan oligosaccharides with various biological activities. [ABSTRACT FROM AUTHOR]

In this paper, a linear piezoelectric micromotor for dual-slider positioning by a single piezoelectric element (PZT) is proposed. A slider 1 and a permanent magnet are connected by the PZT, and a slider 2 is placed on the permanent magnet by the magnetic force. The slider 1 is a small steel cuboid and can be clamped and released by an electromagnet base. When it is released, it can be driven by impact friction force generated by the PZT. When it is clamped, it keeps stationary, and the slider 2 can be positioned based on the smooth impact friction drive of the micromotor. Both the sliders can be positioned independently with a long motion range as well as a high positioning resolution. Due to a single PZT used in the micromotor and miniaturized design of the mechanism, the proposed micromotor has been constructed with a compact size as well as a relatively high loading capacity. [ABSTRACT FROM AUTHOR]

[Display omitted] • Alkaline lignin and DM resins were utilized to modify poplar fast-growing wood. • The materials used were green and environment friendly. • The DM-AL modified wood obviously increased dimensional stability. • The DM-AL modified wood with 12 %AL displayed excellent mechanical properties. • The DM-AL modified wood was found effective against decay fungus. Fast-growing poplar wood was modified by using 1,3-dihydroxymethyl-4,5-dihydroxyethylideneurea (DM) and alkaline lignin (AL) to improve its mechanical properties and decay resistance. AL was first used in combination with DM for wood impregnation modification. The weight percent gain, dimensional stability, mechanical properties, decay resistance and microcosmic changes of all wood samples were evaluated systematically. The results indicated that the maximum weight percent gain and anti-swelling efficiency of poplar wood after modification with DM-AL reached 46.2 % and 52 %, respectively. Compared with the unmodified samples, the bending strength, modulus of elasticity and hardness of the wood samples modified with DM-AL increased by 40 % – 60 %. Notably, the impact strength of samples modified with DM-AL was 18–26 % higher than that of samples modified with DM only, which may be due to the fact that the AL macromolecules slowed down the infiltration of DM-AL into the wood cell walls. The decay resistance of optimal modified samples against the decay fungus was improved by ∼80 % compared with unmodified sample. Furthermore, scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS) confirmed that DM-AL modifiers impregnated into wood cell lumens. Fourier transform infrared spectroscopy (FTIR) demonstrated that the DM resin crosslinked with wood cell walls and AL macromolecules. In general, this study provides a facile, low-cost and green method to improve the comprehensive performance of fast-growing poplar, including mechanical properties and decay resistance. [ABSTRACT FROM AUTHOR]

Features act as a key factor in pedestrian detection task. Most widely-used ones like HOG are manually designed and hard to be adaptive, thus now more attention has been paid to the features automatically learned on data. In this paper, a novel approach of learning discriminative features is proposed, addressing two main limitations of the methods in the literature. On one hand, unlike those methods of learning features on low-level pixels, we propose to learn features via a particular sparse coding algorithm enhanced on mid-level image representation, in order to obtain higher-level latent semantics and robustness; On the other hand, those methods usually utilize label information in model training such as deformable part model (DPM) with high computation cost. Instead, we propose to extend the learning process via a maximum margin criterion, in order to better encode discriminative information directly in features by optimizing them to be close to each other if from the same class and far from each other if from different classes. Furthermore, a boosted detection framework rather than the complex DPM is adopted to achieve both high accuracy and efficiency. The proposed approach achieves promising results on several standard pedestrian detection benchmarks. [ABSTRACT FROM AUTHOR]

Cross-media retrieval is to submit data of any media type, and get semantically relevant results of different media types. Most existing approaches project low-level features of cross-media data onto a unified feature space. However, some of these feature spaces usually have no explicit semantics, which ignore the intrinsic semantic information contained in the original media content. The others only have coarse-grained semantics suffering from the ambiguity of high-level concepts, because the coarse-grained correlation between low-level features and high-level concepts is simply utilized. Hence, the aforementioned approaches cannot generate the descriptive representation of media content, leading to reduced effectiveness to measure the semantic similarities among cross-media data. To address the above problems, we propose a novel approach to cross-media retrieval by exploiting the fine-grained correlation at the entity level and generating the unified descriptive representation. Concretely, the proposed approach first constructs an entity level with fine-grained semantics between low-level features and high-level concepts. Second, by minimizing (maximizing) the distances between media content with positive (negative) correlation at the entity level, we learn the distance-preserving entity projections (DPEP) and generate the unified descriptive representation of media content. Experimental results on two publicly available datasets demonstrate the effectiveness of our approach. [ABSTRACT FROM AUTHOR]

Robust local feature matching plays an important role in the challenging task of logo image recognition. Most traditional methods consider the individual local feature but ignore the affine-invariant geometric relationship among the adjacent local features, which is essential to reduce the number of mismatching. In addition, they do matching for all of the local features and ignore that many ones are insignificant, which increase the probability of mismatching and the computation complexity. To address the two limitations, we propose a robust matching method to get the better matching results by exploiting the distinctive topological constraint together with the feature selection. In the proposed method, first we employ the distinctive topological constraint to enhance the describing ability of local features, which makes full use of the affine-invariant geometric relationship among adjacent local features for more accurate local feature matching. Second, we utilize the feature selection algorithm based on the mutual information (MI), to filter out most insignificant local features before matching, which is efficient and effective to guarantee the performance of local feature matching. We evaluated the proposed method on two challenging datasets, i.e, FlickrLogos-32 and FlickrLogos-27, and achieve superior performance against the state-of-the-art methods in the literature. [ABSTRACT FROM AUTHOR]

In this paper, a novel flapping wing mechanism driven by a linear actuator is proposed. The linear actuator consists of a piezoelectric element and a permanent magnet. Based on the principle of impact drive mechanism, the linear actuator can move a reciprocating linear motion with a long travel range, which can be directly converted to a flapping motion via a crank-slider mechanism. In comparison with conventional flapping wing mechanisms driven by a rotary motor with a crank-rocker mechanism, no gearbox is needed in our design. Therefore, the proposed flapping wing mechanism can be made with small volume and light weight. A prototype was designed and constructed and experiments were carried out to test the performance of the flapping wing mechanism. The experimental results confirm that the designed flapping wing mechanism can obtain a continuous flapping motion by moving the linear actuator reciprocatingly. [ABSTRACT FROM AUTHOR]

With the rapid growth of multimedia data such as text, image, video, audio, and 3-D model, cross-media retrieval has become increasingly important, because users can retrieve the results with various types of media by submitting a query of any media type. Comparing with single-media retrieval such as image retrieval and text retrieval, cross-media retrieval is better because it provides the retrieval results with all kinds of media at the same time. In this paper, we focus on how to learn cross-media features for different media types, which is a key challenge for cross-media retrieval. Existing methods either model different media types separately or only exploit the labeled multimedia data. Actually, the data from different media types with the same semantic category are complementary to each other, and jointly modeling them is able to improve the accuracy of cross-media retrieval. In addition, although the labeled data are accurate, they require a lot of human labor and thus are very scarce. To address the above problems, we propose a semi-supervised cross-media feature learning algorithm with unified patch graph regularization (S ^2 UPG). Our motivation and contribution mainly lie in the following three aspects. First, existing methods only model different media types in different graphs, while we employ one joint graph to simultaneously model all the media types. The joint graph is able to fully exploit the semantic correlations among various media types, which are complementary to provide the rich hint for cross-media correlation. Second, existing methods only consider the original media instances (such as images, videos, texts, audios, and 3-D models) but ignore their patches, while we make full use of both the media instances and their patches in one graph. Cross-media patches could emphasize the important parts and make cross-media correlations more precise. Third, traditional semi-supervised learning methods only exploit single-media unlabeled instances, while our approach fully exploits cross-media unlabeled instances and their patches, which can increase the diversity of training data and boost the accuracy of cross-media retrieval. Comparing with the current state-of-the-art methods on three datasets, including the challenging XMedia dataset with five media types, the comprehensive experimental results show that our proposed approach performs better. [ABSTRACT FROM AUTHOR]

Human activity recognition can be exploited to benefit ubiquitous applications using sensors. Current research on sensor-based activity recognition is mainly using data-driven or knowledge-driven approaches. In terms of complex activity recognition, most data-driven approaches suffer from portability, extensibility and interpretability problems, whilst knowledge-driven approaches are often weak in handling intricate temporal data. To address these issues, we exploit time series shapelets for complex human activity recognition. In this paper, we first describe the association between activity and time series transformed from sensor data. Then, we present a recursively defined multilayered activity model to represent four types of activities and employ a shapelet-based framework to recognize various activities represented in the model. A prototype system was implemented to evaluate our approach on two public datasets. We also conducted two real-world case studies for system evaluation: daily living activity recognition and basketball play activity recognition. The experimental results show that our approach is capable of handling complex activity effectively. The results are interpretable and accurate, and our approach is fast and energy-efficient in real-time. [ABSTRACT FROM AUTHOR]

This paper provides a comprehensive review of the literature regarding precision piezoelectric motors over long ranges based on the principle of repeating a series of small periodic step motions, named “frequency leveraged motors” in this paper. A summary of recent research into frequency leveraged motors is presented. Work is classified into three categories by different frequency driving methods, including ultrasonic motors, quasi-static motors (non-resonant motors), and motors combined resonant and quasi-static operations. Pros and cons of each motor type are discussed in term of their principle, structure, and performance. In addition, future perspectives and improvements of the frequency leveraged motor are also provided. It is summarized in such a way can provide a better understanding of the core characteristics of each type of long range piezoelectric motor. Moreover, it also aids in determining successful designs, suitability for applications and further research areas. [ABSTRACT FROM AUTHOR]

Endophytes can facilitate the improvement of plant growth and health in agriculturally important crops, yet their genomes and secondary metabolites remain largely unexplored. We previously isolated Saccharibacillus brassicae strain ATSA2T from surface-sterilized seeds of kimchi cabbage and represented a novel species of the genus Saccharibacillus. In this study, we evaluated the plant growth-promoting (PGP) effect of strain ATSA2T in kimchi cabbage, bok choy, and pepper plants grown in soils. We found a significant effect on the shoot and root biomass, and chlorophyll contents following strain ATSA2T treatment. Strain ATSA2T displayed PGP traits such as indole acetic acid (IAA, 62.9 μg/mL) and siderophore production, and phosphate solubilization activity. Furthermore, genome analysis of this strain suggested the presence of gene clusters involved in iron acquisition (fhuABD, afuABC, fbpABC, and fepCDG) and phosphate solubilization (pstABCHS, phoABHLU, and phnCDEP) and other phytohormone biosynthesis genes, including indole-3-acetic acid (trpABCDEFG), in the genome. Interestingly, the secondary metabolites cerecidin, carotenoid, siderophore (staphylobactin), and bacillaene underlying plant growth promotion were found in the whole genome via antiSMASH analysis. Overall, physiological testing and genome analysis data provide comprehensive insights into plant growth-promoting mechanisms, suggesting the relevance of strain ATSA2T in agricultural biotechnology. [ABSTRACT FROM AUTHOR]

Cross-media retrieval has become a key problem in both research and application, in which users can search results across all of the media types (text, image, audio, video, and 3-D) by submitting a query of any media type. How to measure the content similarity among different media is the key challenge. Existing cross-media retrieval methods usually focus on modeling the pairwise correlation or semantic information separately. In fact, these two kinds of information are complementary to each other and optimizing them simultaneously can further improve the accuracy. In this paper, we propose a novel feature learning algorithm for cross-media data, called joint representation learning (JRL), which is able to explore jointly the correlation and semantic information in a unified optimization framework. JRL integrates the sparse and semisupervised regularization for different media types into one unified optimization problem, while existing feature learning methods generally focus on a single media type. On one hand, JRL learns sparse projection matrix for different media simultaneously, so different media can align with each other, which is robust to the noise. On the other hand, both the labeled data and unlabeled data of different media types are explored. Unlabeled examples of different media types increase the diversity of training data and boost the performance of joint representation learning. Furthermore, JRL can not only reduce the dimension of the original features, but also incorporate the cross-media correlation into the final representation, which further improves the performance of both cross-media retrieval and single-media retrieval. Experiments on two datasets with up to five media types show the effectiveness of our proposed approach, as compared with the state-of-the-art methods. [ABSTRACT FROM PUBLISHER]

Highlights: [•] A Cr-N thin-film displacement sensor with a millimeter-scale measurement range is proposed for nanometric positioning of a developed micro-stage. [•] The developed sensor has a compact and small size of 14mm (L)×1mm (W)×0.1mm (T). [•] Experiment revealed the resolution of the sensor is approximately 10nm. [•] The sensor can be investigated into precision positioning of micro systems with millimeter-scale operation range. [ABSTRACT FROM AUTHOR]

Highlights: [•] A motion mechanism of XY micro-stage with a millimeter travel range was designed. [•] The developed motion mechanism has a compact apparatus of 24mm×24mm×5mm. [•] A millimeter travel range with a 10nm positioning resolution was achieved. [•] Developed motion mechanism was confirmed to achieve velocity of more than 5mm/s. [ABSTRACT FROM AUTHOR]

With the rapid development of multimedia content on the Internet, cross-media retrieval has become a key problem in both research and application. Cross-media retrieval is able to retrieve the results of the same semantics with the query, but with different media types. For instance, given a query image of Moraine Lake, besides retrieving the images about Moraine Lake, cross-media retrieval system can also retrieve the related media contents of different media types such as text description. As a result, measuring content similarity between different media is a challenging problem. In this paper, we propose a novel cross-media similarity measure. It considers both intra-media and inter-media correlation, which are ignored by existing works. Intra-media correlation focuses on semantic category information within each media, while inter-media correlation focuses on positive and negative correlations between different media types. Both of them are very important and their adaptive fusion can complement each other. To mine the intra-media correlation, we propose a heterogeneous similarity measure with nearest neighbors (HSNN). The heterogeneous similarity is obtained by computing the probability for two media objects belonging to the same semantic category. To mine the inter-media correlation, we propose a cross-media correlation propagation (CMCP) approach to simultaneously deal with positive and negative correlation between media objects of different media types, while existing works focus solely on the positive correlation. Negative correlation is very important because it provides effective exclusive information. The correlations are modeled as must-link constraints and cannot-link constraints, respectively. Furthermore, our approach is able to propagate the correlation between heterogeneous modalities. Finally, both HSNN and CMCP are flexible, so that any traditional similarity measure could be incorporated. An effective ranking model is learned by further fusion of multiple similarity measures through AdaRank for cross-media retrieval. The experimental results on two datasets show the effectiveness of our proposed approach, compared with state-of-the-art methods. [ABSTRACT FROM AUTHOR]

This paper presents a novel pairwise constraint propagation approach by decomposing the challenging constraint propagation problem into a set of independent semi-supervised classification subproblems which can be solved in quadratic time using label propagation based on $$k$$-nearest neighbor graphs. Considering that this time cost is proportional to the number of all possible pairwise constraints, our approach actually provides an efficient solution for exhaustively propagating pairwise constraints throughout the entire dataset. The resulting exhaustive set of propagated pairwise constraints are further used to adjust the similarity matrix for constrained spectral clustering. Other than the traditional constraint propagation on single-source data, our approach is also extended to more challenging constraint propagation on multi-source data where each pairwise constraint is defined over a pair of data points from different sources. This multi-source constraint propagation has an important application to cross-modal multimedia retrieval. Extensive results have shown the superior performance of our approach. [ABSTRACT FROM AUTHOR]

Abstract: This paper proposes a novel latent semantic learning method for extracting high-level latent semantics from a large vocabulary of abundant mid-level features (i.e. visual keywords) with structured sparse representation, which can help to bridge the semantic gap in the challenging task of human action recognition. To discover the manifold structure of mid-level features, we develop a graph-based spectral embedding approach to latent semantic learning, with the graph over mid-level features being constructed using sparse representation. Moreover, we define new hypergraph regularization to induce extra structured sparsity into sparse representation for graph construction. Due to the nice properties (sparsity and noise-robustness) of such structured sparse representation, our graph construction can capture dominant and robust relationships among mid-level features, which are crucial for the success of latent semantic learning in action recognition. Unlike the traditional latent semantic analysis based on topic models, our latent semantic learning method can explore the manifold structure of mid-level features in both graph construction and spectral embedding, which results in compact but discriminative high-level features. The experimental results on the commonly used KTH action dataset and unconstrained YouTube action dataset show the promising performance of our method. [Copyright &y& Elsevier]

Sensitive detection of explosives has always been a frontier research for scientists. In this work, two [3 + 2] imine organic cages (1 and 2) stemmed from distinct TPE-based dialdehydes (TPE-2CHO/TPE-2PCHO) and common tri(2-aminoethyl)amine were obtained. Both of them are fluorescence silence in solution and high-efficient AIE could be observed, which is different from all known TPE-based organic cages with AIEE properties. Because of the cage skeleton encircled by multiple benzene rings, the two cages can effectively detect various explosives including nitrobenzene, 2-nitrotoluene, 3-nitrotoluene, 4-nitrotoluene and 2,6-dinitrotoluene both in solution and in the solid state, in which cage 1 displays higher detection ability due to its nine encircling benzene donors [Display omitted] • Partial benzene ring modified TPE-based dialdehydes were used to construct imine organic cages. • Cages 1 and 2 show typical AIE characteristics. • The two cages can effectively detect various explosives both in water phase and solid state. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents a fast simulated annealing framework for combining multiple clusterings based on agreement measures between partitions, which are originally used to evaluate a clustering algorithm. Although we can follow a greedy strategy to optimize these measures as the objective functions of clustering ensemble, it may suffer from local convergence and simultaneously incur too large computational cost. To avoid local optima, we consider a simulated annealing optimization scheme that operates through single label changes. Moreover, for the measures between partitions based on the relationship (joined or separated) of pairs of objects, we can update them incrementally for each label change, which ensures that our optimization scheme is computationally feasible. The experimental evaluations demonstrate that the proposed framework can achieve promising results. [Copyright &y& Elsevier]

Abstract: When learning Gaussian mixtures from multivariate data, it is crucial to select the appropriate number of components and simultaneously avoid local optima. To resolve these problems, we follow the idea of competitive agglomeration which is originally used for fuzzy clustering and propose two robust algorithms for Gaussian mixture learning. Through some asymptotic analysis, we find that such robust competitive agglomeration can lead to automatic model selection on Gaussian mixtures and also make our algorithms less sensitive to initialization than the EM algorithm. Experiments demonstrate that our algorithms can achieve promising results just as our theoretic analysis. [Copyright &y& Elsevier]

Abstract: This paper presents a novel method to give a good initial estimate of the probabilistic latent semantic analysis (pLSA) model using rival penalized competitive learning (RPCL), since the expectation maximization (EM) algorithm used to train the model is sensitive to the initialization. As a generative model from the statistical text literature, pLSA is further applied to the bag-of-words representation for each image in the database. Especially for those images containing multiple object categories (e.g. grass, roads, and buildings), we aim to discover the objects (i.e., latent topics) in an unsupervised way using pLSA. Based on the discovered topics, image categorization is then carried out by ensemble-based support vector machine (SVM). We then find in the experiments that the pLSA model with RPCL initialization followed by ensemble-based SVM categorization is robust to the changes of the visual vocabulary and the number of latent topics. [Copyright &y& Elsevier]

Lifelong person re-identification (LReID) refers to matching people across different cameras given continuous data streams. The challenge of catastrophic forgetting of old knowledge and the effective acquisition of new knowledge form a significant dilemma for LReID. Most current LReID methods propose to retain abundant exemplars from historical data, which are further rehearsed to fully fine-tune the whole model. However, such a learning paradigm will inevitably hinder data privacy and result in substantial computation costs. In this paper, we propose a paradigm for exemplar-free LReID through model re-parameterization. Without retaining any exemplars, our designed method adopts a novel Prompt-guided Adaptive Exponential Moving Average (PAEMA) strategy to achieve dynamic knowledge consolidation. Our key idea is to leverage visual prompting as the guidance for model re-parameterization to benefit knowledge preservation. Conventional Exponential Moving Average (EMA) methods rely on fixed or time-varied constants as weighting parameters, the unpredictable correlation between new and old data streams may lead to varying levels of model parameter drifting during LReID learning. Hence, we argue that a proper weighting parameter should be conditioned on the variation of new and old models to provide an adaptive knowledge consolidation for LReID. To do so, an adaptive mechanism is proposed to utilize the visual prompt as a surrogate for model variation estimation. Consequently, without using any exemplars, the forgetting issue in LReID is greatly alleviated. Experiments on various LReID benchmarks have verified the superiority of our method against the state-of-the-art lifelong learning and LReID approaches. Code is available at https://github.com/zhoujiahuan1991/IJCV2024-PAEMA/. [ABSTRACT FROM AUTHOR]