Search

Cross-domain object detection aims to transfer knowledge from a labeled dataset to an unlabeled dataset. Most existing methods apply a unified embedding model to generate the tightly coupled source and target descriptions for domain alignment, leading to the destroyed feature distribution of the target domain because the embedding model is mainly controlled by the source domain. To reduce the representation bias of the target domain, we apply two independent networks to extract two types of discriminative descriptions with mutual consistency, i.e., a novel Dual Instance-Consistent Network (DICN) is proposed for cross-domain object detection. Especially, Dual Instance-Consistent Module containing the instance mutual consistency between Primary Network and Auxiliary Network is applied to align two domains, where Primary and Auxiliary Networks are used to obtain the source-specific and target-specific information, respectively. The instance mutual consistency consists of two terms: feature consistency and detection consistency, which is applied to align the instance feature and the output of detection head, respectively. With the instance mutual consistency, optimizing the Primary (Auxiliary) Network only with source (target) images by fixing the Auxiliary (Primary) Network can generate the source(target)-specific description. Extensive experiments on several benchmarks demonstrate the effectiveness of the proposed DICN, e.g., obtaining mAP of 44.10% for Cityscapes → Foggy Cityscapes, AP on car of 76.50% for Cityscapes → KITTI, MR

In this paper, we propose a new task of cross-modal federated human activity recognition (CMF-HAR), which is conducive to promote the large-scale use of the HAR model on more local devices. To address the new task, we propose a feature-disentangled activity recognition network (FDARN), which has five important modules of altruistic encoder, egocentric encoder, shared activity classifier, private activity classifier and modality discriminator. The altruistic encoder aims to collaboratively embed local instances on different clients into a modality-agnostic feature subspace. The egocentric encoder aims to produce modality-specific features that cannot be shared across clients with different modalities. The modality discriminator is used to adversarially guide the parameter learning of the altruistic and egocentric encoders. Through decentralized optimization with a spherical modality discriminative loss, our model can not only generalize well across different clients by leveraging the modality-agnostic features but also capture the modality-specific discriminative characteristics of each client. Extensive experiment results on four datasets demonstrate the effectiveness of our method.

Big progress has been achieved in domain adaptation in decades. Existing works are always based on an ideal assumption that testing target domain are i.i.d. with training target domains. However, due to unpredictable corruptions (e.g., noise and blur) in real data like web images, domain adaptation methods are increasingly required to be corruption robust on target domains. In this paper, we investigate a new task, Corruption-agnostic Robust Domain Adaptation (CRDA): to be accurate on original data and robust against unavailable-for-training corruptions on target domains. This task is non-trivial due to large domain discrepancy and unsupervised target domains. We observe that simple combinations of popular methods of domain adaptation and corruption robustness have sub-optimal CRDA results. We propose a new approach based on two technical insights into CRDA: 1) an easy-to-plug module called Domain Discrepancy Generator (DDG) that generates samples that enlarge domain discrepancy to mimic unpredictable corruptions; 2) a simple but effective teacher-student scheme with contrastive loss to enhance the constraints on target domains. Experiments verify that DDG keeps or even improves performance on original data and achieves better corruption robustness that baselines., Comment: The first literature to investigate the topic of corruption-agnostic robust domain adaptation, a new practical and challenging domain adaptation setting

Facial expression recognition (FER) is a challenging task due to the large appearance variations and the lack of sufficient training data. Conventional deep approaches either learn a good representation through deep models or synthesize images automatically to enlarge the training set. In this paper, we perform both tasks jointly and propose an end-to-end deep model for simultaneous facial expression recognition and facial image synthesis. The proposed model is based on Generative Adversarial Network (GAN) and enjoys several merits. First, the facial image synthesis and facial expression recognition tasks can boost their performance for each other via the unified model. Second, paired images are not required in our facial image synthesis network, which makes the proposed model much more general and flexible. Meanwhile, the generated facial images largely expand the training set and ease the overfitting problem in our FER task. Third, different expressions are encoded in a disentangled manner in a latent space, which enables us to synthesize facial images with arbitrary expressions by exchanging certain parts of their latent identity features. Quantitative and qualitative evaluations on both controlled and in-the-wild FER benchmarks (Multi-PIE, MMI, and RAF-DB) demonstrate the effectiveness of our proposed method on both facial image synthesis and facial expression recognition task.

Video moment retrieval has progressed significantly over the past few years, aiming to search the moment that is most relevant to a given natural language query. Most existing methods are trained in a fully-supervised or a weakly-supervised manner, which requires a time-consuming and expensive manually labeling process. In this work, we propose an alternative approach to achieving video moment retrieval that requires no textual annotations of videos and instead leverages the existing visual concept detectors and a pre-trained image-sentence embedding space. Specifically, we design a video-conditioned sentence generator to produce a suitable sentence representation by utilizing the mined visual concepts in videos. We then design a GNN-based relation-aware moment localizer to reasonably select a portion of video clips under the guidance of the generated sentence. Finally, the pre-trained image-sentence embedding space is adopted to evaluate the matching scores between the generated sentence and moment representations with the knowledge transferred from the image domain. By maximizing these scores, the sentence generator and moment localizer can enhance and complement each other to achieve the moment retrieval task. Experimental results on the Charades-STA and ActivityNet Captions datasets demonstrate the effectiveness of our proposed method.

The β-1,6-glucan is the key linker between mannoproteins in the outermost part of the cell wall and β-1,3-glucan/chitin polysaccharide to maintain the rigid structure of the cell wall. The β-1,6-glucanase GluM, which was purified from the fermentation supernatant of Corallococcus sp. EGB, was able to inhibit the germination of Fusarium oxysporum f. sp. cucumerinum conidia at a minimum concentration of 2.0 U/mL (0.08 μg/mL). The survival rates of GluM-treated conidia and monohyphae were 10.4% and 30.7%, respectively, which were significantly lower than that of β-1,3-glucanase treatment (Zymolyase, 20.0 U/mL; equate to 1.0 mg/mL) (72.9% and 73.9%). In contrast to β-1,3-glucanase treatment, the high-osmolarity glycerol (HOG) pathway of F. oxysporum f. sp. cucumerinum cells was activated after GluM treatment, and the intracellular glycerol content was increased by 2.6-fold. Moreover, the accumulation of reactive oxygen species (ROS) in F. oxysporum f. sp. cucumerinum cells after GluM treatment induced apoptosis, but it was not associated with the increased intracellular glycerol content. Together, the results indicate that β-1,6-glucan is a promising target for the development of novel broad-spectrum antifungal agents. IMPORTANCE Phytopathogenic fungi are the most devastating plant pathogens in agriculture, causing enormous economic losses to global crop production. Biocontrol agents have been promoted as replacements to synthetic chemical pesticides for sustainable agriculture development. Cell wall-degrading enzymes (CWDEs), including chitinases and β-1,3-glucanases, have been considered as important armaments to damage the cell wall. Here, we found that F. oxysporum f. sp. cucumerinum is more sensitive to β-1,6-glucanase GluM treatment (0.08 μg/mL) than β-1,3-glucanase Zymolyase (1.0 mg/mL). The HOG pathway was activated in F. oxysporum f. sp. cucumerinum cells after GluM treatment, and the intracellular glycerol content was significantly increased. Moreover, the decomposition of F. oxysporum f. sp. cucumerinum cell wall by GluM induced the burst of intracellular ROS and apoptosis, which eventually leads to cell death. Therefore, we suggest that the β-1,6-glucan of the fungal cell wall may be a better antifungal target compared to the β-1,3-glucan.

With the growing amount of multimodal data, cross-modal retrieval has attracted more and more attention and become a hot research topic. To date, most of the existing techniques mainly convert multimodal data into a common representation space where similarities in semantics between samples can be easily measured across multiple modalities. However, these approaches may suffer from the following limitations: 1) They overcome the modality gap by introducing loss in the common representation space, which may not be sufficient to eliminate the heterogeneity of various modalities; 2) They treat labels as independent entities and ignore label relationships, which is not conducive to establishing semantic connections across multimodal data; 3) They ignore the non-binary values of label similarity in multi-label scenarios, which may lead to inefficient alignment of representation similarity with label similarity. To tackle these problems, in this paper, we propose two models to learn discriminative and modality-invariant representations for cross-modal retrieval. Firstly, the dual generative adversarial networks are built to project multimodal data into a common representation space. Secondly, to model label relation dependencies and develop inter-dependent classifiers, we employ multi-hop graph neural networks (consisting of Probabilistic GNN and Iterative GNN), where the layer aggregation mechanism is suggested for using propagation information of various hops. Thirdly, we propose a novel soft multi-label contrastive loss for cross-modal retrieval, with the soft positive sampling probability, which can align the representation similarity and the label similarity. Additionally, to adapt to incomplete-modal learning, which can have wider applications, we propose a modal reconstruction mechanism to generate missing features. Extensive experiments on three widely used benchmark datasets, i.e., NUS-WIDE, MIRFlickr, and MS-COCO, show the superiority of our proposed method.

Health management is getting increasing attention all over the world. However, existing health management mainly relies on hospital examination and treatment, which are complicated and untimely. The emergence of mobile devices provides the possibility to manage people’s health status in a convenient and instant way. Estimation of health status can be achieved with various kinds of data streams continuously collected from wearable sensors. However, these data streams are multi-source and heterogeneous, containing complex temporal structures with local contextual and global temporal aspects, which makes the feature learning and data joint utilization challenging. We propose to model the behavior-related multi-source data streams with a local-global graph, which contains multiple local context sub-graphs to learn short-term local context information with heterogeneous graph neural networks and a global temporal sub-graph to learn long-term dependency with self-attention networks. Then health status is predicted based on the structure-aware representation learned from the local-global behavior graph. We take experiments on the StudentLife dataset, and extensive results demonstrate the effectiveness of our proposed model.

Transformers, the dominant architecture for natural language processing, have also recently attracted much attention from computational visual media researchers due to their capacity for long-range representation and high performance. Transformers are sequence-to-sequence models, which use a self-attention mechanism rather than the RNN sequential structure. Thus, such models can be trained in parallel and can represent global information. This study comprehensively surveys recent visual transformer works. We categorize them according to task scenario: backbone design, high-level vision, low-level vision and generation, and multimodal learning. Their key ideas are also analyzed. Differing from previous surveys, we mainly focus on visual transformer methods in low-level vision and generation. The latest works on backbone design are also reviewed in detail. For ease of understanding, we precisely describe the main contributions of the latest works in the form of tables. As well as giving quantitative comparisons, we also present image results for low-level vision and generation tasks. Computational costs and source code links for various important works are also given in this survey to assist further development.

Lithium-rich manganese-based cathode materials have become one of the most concerned cathode materials for high-energy lithium-ion batteries. In order to improve its electrochemical performance, Li1.2Mn0.54Ni0.13Co0.13O2 with different content LiNbO3 coatings was synthesized by mechanical ball milling. The morphology, microstructure, and electrochemical properties of the samples were investigated by X-ray diffraction, scanning electron microscope, transmission electron microscope, galvanostatic charge/discharge, electrochemical impedance spectroscopy, and cyclic voltammetry. The results show that LiNbO3 coating not only protects the cathode material from the corrosion of electrolyte and HF but also improves the migration rate of Li+ in the interface region. Notably, the 5 wt% LiNbO3-coated Li1.2Mn0.54Ni0.13Co0.13O2 exhibits capacity retention of 89.9% under 0.1 C after 100 cycles. Besides, it has a higher discharge capacity than Li1.2Mn0.54Ni0.13Co0.13O2 at different rates. LiNbO3 coating is an effective way to improve its cycle stability and rate performance of Li1.2Mn0.54Ni0.13Co0.13O2.

Point-level weakly-supervised temporal action localization (P-WSTAL) aims to localize temporal extents of action instances and identify the corresponding categories with only a single point label for each action instance for training. Due to the sparse frame-level annotations, most existing models are in the localization-by-classification pipeline. However, there exist two major issues in this pipeline: large intra-action variation due to task gap between classification and localization and noisy classification learning caused by unreliable pseudo training samples. In this paper, we propose a novel framework CRRC-Net, which introduces a co-supervised feature learning module and a probabilistic pseudo label mining module, to simultaneously address the above two issues. Specifically, the co-supervised feature learning module is applied to exploit the complementary information in different modalities for learning more compact feature representations. Furthermore, the probabilistic pseudo label mining module utilizes the feature distances from action prototypes to estimate the likelihood of pseudo samples and rectify their corresponding labels for more reliable classification learning. Comprehensive experiments are conducted on different benchmarks and the experimental results show that our method achieves favorable performance with the state-of-the-art.

Digital art synthesis is receiving increasing attention in the multimedia community because of engaging the public with art effectively. Current digital art synthesis methods usually use single-modality inputs as guidance, thereby limiting the expressiveness of the model and the diversity of generated results. To solve this problem, we propose the multimodal guided artwork diffusion (MGAD) model, which is a diffusion-based digital artwork generation approach that utilizes multimodal prompts as guidance to control the classifier-free diffusion model. Additionally, the contrastive language-image pretraining (CLIP) model is used to unify text and image modalities. Extensive experimental results on the quality and quantity of the generated digital art paintings confirm the effectiveness of the combination of the diffusion model and multimodal guidance. Code is available at https://github.com/haha-lisa/MGAD-multimodal-guided-artwork-diffusion., Accepted by ACM MM 2022

Multi-target multi-camera tracking (MTMCT) targets to generate trajectories of the object that appeared under multiple cameras automatically. MTMCT can be treated as a combination of intra-camera tracking and cross-camera tracking. The existing work only employs the global description to perform the tracklet generating. However, the global description cannot model the local similarity between targets, leading to existing methods not to be robust to occlusion and fast motion. To handle the mentioned problem, we propose an online Optical-based Pose Association (OPA) for multi-target multi-camera tracking. The proposed method utilizes local pose matching to solve the occlusion problem, and applies optical flow to reduce the distance caused by fast motion. For optical-based pose association, we firstly employ OpenPose to generate human pose for each proposal. Then, we utilize the optical flow generated by PWC-Net to adjust the estimated pose for the previous frame. Finally, the modified Object Keypoint Similarity is used to compute the similarity between the pose of the current frame and adjusted pose in the prior frame. Once obtaining the optical-based pose similarity, we combine it with the visual and bounding box spatial similarities to generate the final similarity matrix, and apply the Kuhn-Munkras algorithm for data association. The experiments on the MTMCT and MOT datasets verify the rationality of using human pose information and prove the superiority of the proposed method.

We target at the task of weakly-supervised video object grounding (WSVOG), where only video-sentence annotations are available during model learning. It aims to localize objects described in the sentence to visual regions in the video, which is a fundamental capability needed in pattern analysis and machine learning. Despite the recent progress, existing methods all suffer from the severe problem of spurious association, which will harm the grounding performance. In this paper, we start from the definition of WSVOG and pinpoint the spurious association from two aspects: (1) the association itself is not object-relevant but extremely ambiguous due to weak supervision, and (2) the association is unavoidably confounded by the observational bias when taking the statistics-based matching strategy in existing methods. With this in mind, we design a unified causal framework to learn the deconfounded object-relevant association for more accurate and robust video object grounding. Specifically, we learn the object-relevant association by causal intervention from the perspective of video data generation process. To overcome the problems of lacking fine-grained supervision in terms of intervention, we propose a novel spatial-temporal adversarial contrastive learning paradigm. To further remove the accompanying confounding effect within the object-relevant association, we pursue the true causality by conducting causal intervention via backdoor adjustment. Finally, the deconfounded object-relevant association is learned and optimized under a unified causal framework in an end-to-end manner. Extensive experiments on both IID and OOD testing sets of three benchmarks demonstrate its accurate and robust grounding performance against state-of-the-arts.

Semi-supervised object detection has recently achieved substantial progress. As a mainstream solution, the self-labeling-based methods train the detector on both labeled data and unlabeled data with pseudo labels predicted by the detector itself, but their performances are always limited. Through experimental analysis, we reveal the underlying reason is that the detector is misguided by the incorrect pseudo labels predicted by itself (dubbed self-errors). These self-errors can hurt performance even worse than random-errors, and can be neither discerned nor rectified during the self-labeling process. In this paper, we propose an effective detection framework named CrossRectify, to obtain accurate pseudo labels by simultaneously training two detectors with different initial parameters. Specifically, the proposed approach leverages the disagreements between detectors to discern the self-errors and refines the pseudo label quality by the proposed cross-rectifying mechanism. Extensive experiments show that CrossRectify achieves outperforming performances over various detector structures on 2D and 3D detection benchmarks.

Background/aimsTo improve the accuracy of pterygium screening and detection through smartphones, we established a fusion training model by blending a large number of slit-lamp image data with a small proportion of smartphone data.MethodTwo datasets were used, a slit-lamp image dataset containing 20 987 images and a smartphone-based image dataset containing 1094 images. The RFRC (Faster RCNN based on ResNet101) model for the detection model. The SRU-Net (U-Net based on SE-ResNeXt50) for the segmentation models. The open-cv algorithm measured the width, length and area of pterygium in the cornea.ResultsThe detection model (trained by slit-lamp images) obtained the mean accuracy of 95.24%. The fusion segmentation model (trained by smartphone and slit-lamp images) achieved a microaverage F1score of 0.8981, sensitivity of 0.8709, specificity of 0.9668 and area under the curve (AUC) of 0.9295. Compared with the same group of patients’ smartphone and slit-lamp images, the fusion model performance in smartphone-based images (F1score of 0.9313, sensitivity of 0.9360, specificity of 0.9613, AUC of 0.9426, accuracy of 92.38%) is close to the model (trained by slit-lamp images) in slit-lamp images (F1score of 0.9448, sensitivity of 0.9165, specificity of 0.9689, AUC of 0.9569 and accuracy of 94.29%).ConclusionOur fusion model method got high pterygium detection and grading accuracy in insufficient smartphone data, and its performance is comparable to experienced ophthalmologists and works well in different smartphone brands.

We target at the task of weakly-supervised action localization (WSAL), where only video-level action labels are available during model training. Despite the recent progress, existing methods mainly embrace a localization-by-classification paradigm and overlook the fruitful fine-grained temporal distinctions between video sequences, thus suffering from severe ambiguity in classification learning and classification-to-localization adaption. This paper argues that learning by contextually comparing sequence-to-sequence distinctions offers an essential inductive bias in WSAL and helps identify coherent action instances. Specifically, under a differentiable dynamic programming formulation, two complementary contrastive objectives are designed, including Fine-grained Sequence Distance (FSD) contrasting and Longest Common Subsequence (LCS) contrasting, where the first one considers the relations of various action/background proposals by using match, insert, and delete operators and the second one mines the longest common subsequences between two videos. Both contrasting modules can enhance each other and jointly enjoy the merits of discriminative action-background separation and alleviated task gap between classification and localization. Extensive experiments show that our method achieves state-of-the-art performance on two popular benchmarks. Our code is available at https://github.com/MengyuanChen21/CVPR2022-FTCL., Accepted by CVPR 2022. Code is available at https://github.com/MengyuanChen21/CVPR2022-FTCL

Pretrained vision-language models (VLMs) such as CLIP have shown impressive generalization capability in downstream vision tasks with appropriate text prompts. Instead of designing prompts manually, Context Optimization (CoOp) has been recently proposed to learn continuous prompts using taskspecific training data. Despite the performance improvements on downstream tasks, several studies have reported that CoOp suffers from the overfitting issue in two aspects: (i) the test accuracy on base classes first improves and then worsens during training;(ii) the test accuracy on novel classes keeps decreasing. However, none of the existing studies can understand and mitigate such overfitting problems. In this study, we first explore the cause of overfitting by analyzing the gradient flow. Comparative experiments reveal that CoOp favors generalizable and spurious features in the early and later training stages, respectively, leading to the non-overfitting and overfitting phenomena. Given those observations, we propose Subspace Prompt Tuning (SubPT) to project the gradients in back-propagation onto the low-rank subspace spanned by the early-stage gradient flow eigenvectors during the entire training process and successfully eliminate the overfitting problem. In addition, we equip CoOp with a Novel Feature Learner (NFL) to enhance the generalization ability of the learned prompts onto novel categories beyond the training set, needless of image training data. Extensive experiments on 11 classification datasets demonstrate that SubPT+NFL consistently boost the performance of CoOp and outperform the state-of-the-art CoCoOp approach. Experiments on more challenging vision downstream tasks, including open-vocabulary object detection and zero-shot semantic segmentation, also verify the effectiveness of the proposed method. Codes can be found at https://tinyurl.com/mpe64f89.

MicroRNAs (miRNAs or miRs) play important roles in cardiovascular disease. miR-21-5p is known to be involved in the regulation of cardiomyocyte injury under high glucose and high fat (HG-HF) conditions, but its mechanism of action remains unclear. In the present study, a cardiomyocyte cell line, H9c2, was treated with 33 mM glucose and 250 µM sodium palmitate for 24, 48, and 72 h to produce HG-HF injury. After treatment, miR-21-5p expression was detected by reverse transcription-quantitative PCR. A miR-21-5p mimic was then constructed and transfected into the cells and the potential molecular mechanism was investigated using Cell Counting Kit-8, TUNEL, flow cytometry and western blot assays. Expression of miR-21-5p was significantly downregulated by HG-HF treatment of H9c2 cells for 24, 48, and 72 h. In subsequent experiments, cells were treated for an intermediate period (48 h). Compared with the control group, HG-HF treatment significantly inhibited H9c2 proliferation and promoted apoptosis, while these effects were significantly reduced in the miR-21-5p mimic. Compared with the control group, HG-HF treatment significantly increased reactive oxygen species, while miR-21-5p mimic significantly reduced this effect. Compared with the control group, HG-HF treatment significantly increased the expression of the pro-apoptotic proteins Bax and phosphorylated (p)-Akt and decreased the expression of the anti-apoptotic proteins Bcl-2, p-PTEN, and p-FOXO3a, while overexpression of miR-21-5p significantly reduced these effects. The results revealed that miR-21-5p inhibited apoptosis and oxidative stress in H9c2 cells induced by HG-HF, likely through the PTEN/Akt/FOXO3a signaling pathway.

Composed Query Based Image Retrieval (CQBIR) aims at retrieving images relevant to a composed query containing a reference image with a requested modification expressed via a textual sentence. Compared with the conventional image retrieval which takes one modality as query to retrieve relevant data of another modality, CQBIR poses great challenge over the semantic gap between the reference image and modification text in the composed query. To solve the challenge, previous methods either resort to feature composition that cannot model interactions in the query or explore inter-modal attention while ignoring the spatial structure and visual-semantic relationship. In this paper, we propose a geometry sensitive cross-modal reasoning network for CQBIR by jointly modeling the geometric information of the image and the visual-semantic relationship between the reference image and modification text in the query. Specifically, it contains two key components: a geometry sensitive inter-modal attention module (GS-IMA) and a text-guided visual reasoning module (TG-VR). The GS-IMA introduces the spatial structure into the inter-modal attention in both implicit and explicit manners. The TG-VR models the unequal semantics not included in the reference image to guide further visual reasoning. As a result, our method can learn effective feature for the composed query which does not exhibit literal alignment. Comprehensive experimental results on three standard benchmarks demonstrate that the proposed model performs favorably against state-of-the-art methods.

We introduce tf_geometric, an efficient and friendly library for graph deep learning, which is compatible with both TensorFlow 1.x and 2.x. tf_geometric provides kernel libraries for building Graph Neural Networks (GNNs) as well as implementations of popular GNNs. The kernel libraries consist of infrastructures for building efficient GNNs, including graph data structures, graph map-reduce framework, graph mini-batch strategy, etc. These infrastructures enable tf_geometric to support single-graph computation, multi-graph computation, graph mini-batch, distributed training, etc.; therefore, tf_geometric can be used for a variety of graph deep learning tasks, such as transductive node classification, inductive node classification, link prediction, and graph classification. Based on the kernel libraries, tf_geometric implements a variety of popular GNN models for different tasks. To facilitate the implementation of GNNs, tf_geometric also provides some other libraries for dataset management, graph sampling, etc. Different from existing popular GNN libraries, tf_geometric provides not only Object-Oriented Programming (OOP) APIs, but also Functional APIs, which enable tf_geometric to handle advanced graph deep learning tasks such as graph meta-learning. The APIs of tf_geometric are friendly, and they are suitable for both beginners and experts. In this paper, we first present an overview of tf_geometric's framework. Then, we conduct experiments on some benchmark datasets and report the performance of several popular GNN models implemented by tf_geometric., Comment: 7 pages, 5 figures

Splitting a droplet into several segments is of great significance in many applications such as the detection of tiny liquid samples, whereas the surface tension tends to hold liquid to remain as one drop, causing difficulty in separating the droplet into pieces. In this work, a method is proposed to split an impacting droplet with a relatively high velocity or Weber number into two halves by a superhydrophobic wire. The effects of the wire wettability and the impact velocity of the droplet on the splitting phenomena and the efficacy to an anti-icing application are investigated. Compared to a hydrophilic wire, a superhydrophobic wire splits an impacting droplet at a relatively high speed of the Weber number greater than 3.1 and inhibits ice accretion at the temperature as low as −20 °C. The results suggest that a superhydrophobic wire can be utilized in the droplet manipulation and anti-icing applications such as power lines in high latitude areas.

We use the overlapping generations model (OLG) to analyze and simulate the optimal retirement decision based on the two criteria of increasing the sustainability of the Social Security Fund (SSF) and improving the welfare of delayed retirees. Our key findings are as follows: (1) Relative to normal retirement, delayed retirement policies are preferable, and voluntary delayed retirement is preferable to passive delayed retirement. (2) Increases in longevity and labor productivity enhance the policy effects of voluntary delayed retirement, while increases in the birth rate promote fund sustainability but weaken its welfare effects. (3) Therefore, when any of these factors increase simultaneously, the sustainability effect is enhanced by superposition, while the birth rate causes the welfare effect to be weakened by superposition. (4) In addition, the result that pensions would crowd out savings and longevity would lead to greater pressure on pension payments does not necessarily hold.

Vision transformers (ViTs) have recently received explosive popularity, but the huge computational cost is still a severe issue. Since the computation complexity of ViT is quadratic with respect to the input sequence length, a mainstream paradigm for computation reduction is to reduce the number of tokens. Existing designs include structured spatial compression that uses a progressive shrinking pyramid to reduce the computations of large feature maps, and unstructured token pruning that dynamically drops redundant tokens. However, the limitation of existing token pruning lies in two folds: 1) the incomplete spatial structure caused by pruning is not compatible with structured spatial compression that is commonly used in modern deep-narrow transformers; 2) it usually requires a time-consuming pre-training procedure. To tackle the limitations and expand the applicable scenario of token pruning, we present Evo-ViT, a self-motivated slow-fast token evolution approach for vision transformers. Specifically, we conduct unstructured instance-wise token selection by taking advantage of the simple and effective global class attention that is native to vision transformers. Then, we propose to update the selected informative tokens and uninformative tokens with different computation paths, namely, slow-fast updating. Since slow-fast updating mechanism maintains the spatial structure and information flow, Evo-ViT can accelerate vanilla transformers of both flat and deep-narrow structures from the very beginning of the training process. Experimental results demonstrate that our method significantly reduces the computational cost of vision transformers while maintaining comparable performance on image classification., We propose a novel and effective design for dynamic vision transformer to achieve better computational efficiency. The code is available at https://github.com/YifanXu74/Evo-ViT