Search

In the field of Internet-of-Things (IoT), how to enhance the computation capacity and prolong the life span of massive smart IoT nodes is becoming a more practically important but challenging problem. In this paper, we propose a backscatter-assisted wireless-powered non-orthogonal multiple access (NOMA)-mobile edge computing (MEC) network, where IoT nodes harvest energy from signals transmitted by a dedicated energy source and then offload data to the MEC server via backscattering and uplink NOMA. To exploit its potential performance, we formulate a non-convex problem to maximize the total computation bits of all the IoT nodes by jointly optimizing the communication resource of each IoT node and the computation resource of both the IoT nodes and the MEC server. By introducing a slack variable and a series of auxiliary variables as well as using the contradiction and inequality transformation approaches, we transform the original non-convex problem into a convex one and devise an iterative algorithm to optimally solve it. Simulation results are provided to demonstrate the quick convergence of the proposed iterative algorithm and to confirm the superiority of the proposed scheme over the existing benchmark schemes. [ABSTRACT FROM AUTHOR]

Influential node identification plays an important role in optimizing network structure. Many measures and identification methods are proposed for this purpose. However, the current network system is more complex, the existing methods are difficult to deal with these networks. In this paper, several basic measures are introduced and discussed and we propose an improved influential nodes identification method that adopts the hybrid mechanism of information entropy and weighted degree of edge to improve the accuracy of identification (Hm-shell). Our proposed method is evaluated by comparing with nine algorithms in nine datasets. Theoretical analysis and experimental results on real datasets show that our method outperforms other methods on performance. [ABSTRACT FROM AUTHOR]

As an emerging data modal with precise distance sensing, LiDAR point clouds have been placed great expectations on 3D scene understanding. However, point clouds are always sparsely distributed in the 3D space, and with unstructured storage, which makes it difficult to represent them for effective 3D object detection. To this end, in this work, we regard point clouds as hollow-3D data and propose a new architecture, namely Hallucinated Hollow-3D R-CNN (H23D R-CNN), to address the problem of 3D object detection. In our approach, we first extract the multi-view features by sequentially projecting the point clouds into the perspective view and the bird-eye view. Then, we hallucinate the 3D representation by a novel bilaterally guided multi-view fusion block. Finally, the 3D objects are detected via a box refinement module with a novel Hierarchical Voxel RoI Pooling operation. The proposed H23D R-CNN provides a new angle to take full advantage of complementary information in the perspective view and the bird-eye view with an efficient framework. We evaluate our approach on the public KITTI Dataset and Waymo Open Dataset. Extensive experiments demonstrate the superiority of our method over the state-of-the-art algorithms with respect to both effectiveness and efficiency. The code is available at https://github.com/djiajunustc/H-23D_R-CNN. [ABSTRACT FROM AUTHOR]

Color demosaicing is an essential step in the camera image processing pipeline, especially on hardware platforms for real-time video applications. Though many demosaicing algorithms have been proposed in the past two decades, there remains a substantial gap between industrial needs and academic research. On one hand, industry requires a high perceptual quality, artifact-free, and low-cost demosaicing algorithm that is noniterative and small window based ready to be implemented on a hardware platform with a limited line buffer. On the other hand, academia is targeting high PSNR/SSIM, with the computation cost and line buffer receiving second priority, and often a frame buffer and iterative operation are used. The cost, large line buffer and iteration requirement make most existing demosaicing algorithms inapplicable on hardware platforms. In this paper, we introduce a novel low-cost demosaicing algorithm to narrow the gap. We keep the operation window size and computation cost as the first priority and achieve both high PSNR/SSIM and visual perceptual quality compared to previous state-of-the-art methods on standard test datasets. We fully investigate the a priori knowledge of natural scene raw images and find several key cues that are beneficial to demosaicing. Our demosaicing algorithm is a smart fusion of these useful cues. Furthermore, we solve typical demosaicing issues that occur in many traditional methods, including false color artifacts, T-section closing, and zippering artifacts. The proposed method has no learning stage, no iteration operation, a small line buffer and a limited number of parameters, so it can easily be applied to hardware platforms. [ABSTRACT FROM AUTHOR]

In visual tracking, how to effectively model the target appearance using limited prior information remains an open problem. In this paper, we leverage an ensemble of diverse models to learn manifold representations for robust object tracking. The proposed ensemble framework includes a shared backbone network for efficient feature extraction and multiple head networks for independent predictions. Trained by the shared data within an identical structure, the mutually correlated head models heavily hinder the potential of ensemble learning. To shrink the representational overlaps among multiple models while encouraging the diversity of individual predictions, we propose the model diversity and response diversity regularization terms during training. By fusing these distinctive prediction results via a fusion module, the tracking variance caused by the distractor objects can be largely restrained. Our whole framework is end-to-end trained in a data-driven manner, avoiding the heuristic designs of multiple base models and fusion strategies. The proposed method achieves state-of-the-art results on seven challenging benchmarks while operating in real-time. [ABSTRACT FROM AUTHOR]

Person re-identification is a crucial task of identifying pedestrians of interest across multiple surveillance camera views. For person re-identification, a pedestrian is usually represented with features extracted from a rectangular image region that inevitably contains the scene background, which incurs ambiguity to distinguish different pedestrians and degrades the accuracy. Thus, we propose an end-to-end foreground-aware network to discriminate against the foreground from the background by learning a soft mask for person re-identification. In our method, in addition to the pedestrian ID as supervision for the foreground, we introduce the camera ID of each pedestrian image for background modeling. The foreground branch and the background branch are optimized collaboratively. By presenting a target attention loss, the pedestrian features extracted from the foreground branch become more insensitive to backgrounds, which greatly reduces the negative impact of changing backgrounds on pedestrian matching across different camera views. Notably, in contrast to existing methods, our approach does not require an additional dataset to train a human landmark detector or a segmentation model for locating the background regions. The experimental results conducted on three challenging datasets, i.e., Market-1501, DukeMTMC-reID, and MSMT17, demonstrate the effectiveness of our approach. [ABSTRACT FROM AUTHOR]

Cross-modal retrieval aims to identify relevant data across different modalities. In this work, we are dedicated to cross-modal retrieval between images and text sentences, which is formulated into similarity measurement for each image-text pair. To this end, we propose a Cross-modal Relation Guided Network (CRGN) to embed image and text into a latent feature space. The CRGN model uses GRU to extract text feature and ResNet model to learn the globally guided image feature. Based on the global feature guiding and sentence generation learning, the relation between image regions can be modeled. The final image embedding is generated by a relation embedding module with an attention mechanism. With the image embeddings and text embeddings, we conduct cross-modal retrieval based on the cosine similarity. The learned embedding space well captures the inherent relevance between image and text. We evaluate our approach with extensive experiments on two public benchmark datasets, i.e., MS-COCO and Flickr30K. Experimental results demonstrate that our approach achieves better or comparable performance with the state-of-the-art methods with notable efficiency. [ABSTRACT FROM AUTHOR]

Visual can be easily disturbed by similar surrounding objects. Such objects as hard distractors, even though being the minority among negative samples, increase the risk of target drift and model corruption, which deserve additional attention in online tracking and model update. To enhance the tracking robustness, in this paper, we propose a cascaded regression tracker with two sequential stages. In the first stage, we filter out abundant easily-identified negative candidates via an efficient convolutional regression. In the second stage, a discrete sampling based ridge regression is designed to double-check the remaining ambiguous hard samples, which serves as an alternative of fully-connected layers and benefits from the closed-form solver for efficient learning. During the model update, we utilize the hard negative mining technique and an adaptive ridge regression scheme to improve the discrimination capability of the second-stage regressor. Extensive experiments are conducted on 11 challenging tracking benchmarks including OTB-2013, OTB-2015, VOT2018, VOT2019, UAV123, Temple-Color, NfS, TrackingNet, LaSOT, UAV20L, and OxUvA. The proposed method achieves state-of-the-art performance on prevalent benchmarks, while running in a real-time speed. [ABSTRACT FROM AUTHOR]

The advancement of visual tracking has continuously been brought by deep learning models. Typically, supervised learning is employed to train these models with expensive labeled data. In order to reduce the workload of manual annotation and learn to track arbitrary objects, we propose an unsupervised learning method for visual tracking. The motivation of our unsupervised learning is that a robust tracker should be effective in bidirectional tracking. Specifically, the tracker is able to forward localize a target object in successive frames and backtrace to its initial position in the first frame. Based on such a motivation, in the training process, we measure the consistency between forward and backward trajectories to learn a robust tracker from scratch merely using unlabeled videos. We build our framework on a Siamese correlation filter network, and propose a multi-frame validation scheme and a cost-sensitive loss to facilitate unsupervised learning. Without bells and whistles, the proposed unsupervised tracker achieves the baseline accuracy of classic fully supervised trackers while achieving a real-time speed. Furthermore, our unsupervised framework exhibits a potential in leveraging more unlabeled or weakly labeled data to further improve the tracking accuracy. [ABSTRACT FROM AUTHOR]

Void nucleation, growth, and coalescence at extreme strain rates in ductile metals with weak mechanical anisotropies, e.g., copper, iron, and aluminum, have been extensively investigated. However, the atomic-scale fracture properties of strongly anisotropic metals, especially hexagonal close-packed (HCP) metals, at ultrahigh strain rates have rarely been studied. We have investigated the nucleation, growth, and coalescence of voids in HCP-Zr under isotropic-triaxial tension using molecular dynamics (MD) and void nucleation and growth (NAG) models. The effects of temperature were also examined by MD. The void evolution predicted by MD corresponded to that predicted by the NAG model and is divided into three stages, i.e., an initial nucleation stage, an exponential growth stage, and a linear stage. The nucleation threshold Pn0 is very sensitive to temperature, while the growth threshold Pg0 decreases slightly with increasing temperature. The initial NAG parameters were evaluated by an improved optimized genetic algorithm. In addition, we adjusted the NAG parameters until the history of the void volume fraction calculated by these parameters was exactly the same as that calculated by MD. This study predicts comprehensive NAG parameters for HCP-Zr under extreme conditions, providing a valuable reference for future studies of dynamic damage in HCP materials. [ABSTRACT FROM AUTHOR]

Correlation filters (CF) have received considerable attention in visual tracking because of their computational efficiency. Leveraging deep features via off-the-shelf CNN models (e.g., VGG), CF trackers achieve state-of-the-art performance while consuming a large number of computing resources. This limits deep CF trackers to be deployed to many mobile platforms on which only a single-core CPU is available. In this paper, we propose to jointly compress and transfer off-the-shelf CNN models within a knowledge distillation framework. We formulate a CNN model pretrained from the image classification task as a teacher network, and distill this teacher network into a lightweight student network as the feature extractor to speed up CF trackers. In the distillation process, we propose a fidelity loss to enable the student network to maintain the representation capability of the teacher network. Meanwhile, we design a tracking loss to adapt the objective of the student network from object recognition to visual tracking. The distillation process is performed offline on multiple layers and adaptively updates the student network using a background-aware online learning scheme. The online adaptation stage exploits the background contents to improve the feature discrimination of the student network. Extensive experiments on six standard datasets demonstrate that the lightweight student network accelerates the speed of state-of-the-art deep CF trackers to real-time on a single-core CPU while maintaining almost the same tracking accuracy. [ABSTRACT FROM AUTHOR]

Vision-based sign language translation (SLT) is a challenging task due to the complicated variations of facial expressions, gestures, and articulated poses involved in sign linguistics. As a weakly supervised sequence-to-sequence learning problem, in SLT there are usually no exact temporal boundaries of actions. To adequately explore temporal hints in videos, we propose a novel framework named Hierarchical deep Recurrent Fusion (HRF). Aiming at modeling discriminative action patterns, in HRF we design an adaptive temporal encoder to capture crucial RGB visemes and skeleton signees. Specifically, RGB visemes and skeleton signees are learned by the same scheme named Adaptive Clip Summarization (ACS), respectively. ACS consists of three key modules, i.e., variable-length clip mining, adaptive temporal pooling, and attention-aware weighting. Besides, based on unaligned action patterns (RGB visemes and skeleton signees), a query-adaptive decoding fusion is proposed to translate the target sentence. Extensive experiments demonstrate the effectiveness of the proposed HRF framework. [ABSTRACT FROM AUTHOR]

Sign language recognition (SLR) is an important and challenging research topic in the multimedia field. Conventional techniques for SLR rely on hand-crafted features, which achieve limited success. In this paper, we present attention-based 3D-convolutional neural networks (3D-CNNs) for SLR. The framework has two advantages: 3D-CNNs learn spatio-temporal features from raw video without prior knowledge and the attention mechanism helps to select the clue. When training 3D-CNN for capturing spatio-temporal features, spatial attention is incorporated into the network to focus on the areas of interest. After feature extraction, temporal attention is utilized to select the significant motions for classification. The proposed method is evaluated on two large scale sign language data sets. The first one, collected by ourselves, is a Chinese sign language data set that consists of 500 categories. The other is the ChaLearn14 benchmark. The experiment results demonstrate the effectiveness of our approach compared with state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

In content-based image retrieval, SIFT feature and the feature from deep convolutional neural network (CNN) have demonstrated promising performance. To fully explore both visual features in a unified framework for effective and efficient retrieval, we propose a collaborative index embedding method to implicitly integrate the index matrices of them. We formulate the index embedding as an optimization problem from the perspective of neighborhood sharing and solve it with an alternating index update scheme. After the iterative embedding, only the embedded CNN index is kept for on-line query, which demonstrates significant gain in retrieval accuracy, with very economical memory cost. Extensive experiments have been conducted on the public datasets with million-scale distractor images. The experimental results reveal that, compared with the recent state-of-the-art retrieval algorithms, our approach achieves competitive accuracy performance with less memory overhead and efficient query computation. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the problem of scalable visual feature matching in large-scale image search and propose a novel cascaded scalar quantization scheme in dual resolution. We formulate the visual feature matching as a range-based neighbor search problem and approach it by identifying hyper-cubes with a dual-resolution scalar quantization strategy. Specifically, for each dimension of the PCA-transformed feature, scalar quantization is performed at both coarse and fine resolutions. The scalar quantization results at the coarse resolution are cascaded over multiple dimensions to index an image database. The scalar quantization results over multiple dimensions at the fine resolution are concatenated into a binary super-vector and stored into the index list for efficient verification. The proposed cascaded scalar quantization (CSQ) method is free of the costly visual codebook training and thus is independent of any image descriptor training set. The index structure of the CSQ is flexible enough to accommodate new image features and scalable to index large-scale image database. We evaluate our approach on the public benchmark datasets for large-scale image retrieval. Experimental results demonstrate the competitive retrieval performance of the proposed method compared with several recent retrieval algorithms on feature quantization. [ABSTRACT FROM AUTHOR]

This paper presented an adaptive spatial information fuzzy clustering algorithm to improve the image segmentation effect. The algorithm combined spatial information of image and FCM algorithm, which improved the objective function of FCM. At the same time, it used information maximization to identify noise data and eliminated outliers. Experiment results on compo-site images and MRI brain image database Brainweb show that the presented algorithm can segment image adaptively and identify noise data effectively. The algorithm can also solve the lack of spatial information using FCM to segment images, improve algorithm robustness. The proposed algorithm has better image segmentation effect than several advanced clustering algorithms. [ABSTRACT FROM AUTHOR]

For vehicle to vehicle communication (V2VC) system, due to the high mobility, dynamic network topology and limited network resources, resulting in a communication network congestion problem, this thesis proposed a novel distributed beacon scheduling scheme. It called context-aware beacon scheduling (CABS), which was based on spatial context information similar to TDMA transmission with dynamic scheduling beacon, and transmitted beacon under highly reliable and predictable delays. Based on partial measurement of channel load and context information included in the beacon, each node dynamically scheduled beacon transmission in a distributed manner in the allocated time slot. In the simulation, it used various communication scenarios in the real channel model and IEEE 802.11p PHY/MAC model to evaluate the proposed beacon scheduling scheme. The simulation results show that the performance of CABS in packet transmission rate and channel access delay is better than periodic scheduling. Moreover, CABS scheme meets the needs of security applications. [ABSTRACT FROM AUTHOR]

Recently, great advance has been made in large-scale content-based image search. Most state-of-the-art approaches are based on the bag-of-visual-words model with local features, such as SIFT, for image representation. Visual matching between images is obtained by vector quantization of local features. Feature quantization is either performed with hierarchical k-NN which introduces severe quantization loss, or with ANN (approximate nearest neighbors) search such as k-d tree, which is computationally inefficient. Besides, feature matching by quantization ignores the vector distance between features, which may cause many false-positive matches. In this paper, we propose constructing a supporting visual word table for all visual words by visual word expansion. Given the initial quantization result, multiple approximate nearest visual words are identified by checking supporting visual word table, which benefits the retrieval recall. Moreover, we present a matching verification scheme based on binary SIFT (BSIFT) signature. The L distance between original SIFT descriptors is demonstrated to be well kept with the metric of Hamming distance between the corresponding binary SIFT signatures. With the BSIFT verification, false-positive matches can be effectively and efficiently identified and removed, which greatly improves the precision of large-scale image search. We evaluate the proposed approach on two public datasets for large-scale image search. The experimental results demonstrate the effectiveness and efficiency of our scheme. [ABSTRACT FROM AUTHOR]

Alternative splicing enhances transcriptome diversity in all eukaryotes and plays a role in plant tissue identity and stress adaptation. To catalog new maize (Zea mays) transcripts and identify genomic loci that regulate alternative splicing, we analyzed over 90 RNA-seq libraries from maize inbred lines B73 and Mo17, as well as Syn10 doubled haploid lines (progenies from B73 × Mo17). Transcript discovery was augmented with publicly available data from 14 maize tissues, expanding the maize transcriptome by more than 30,000 and increasing the percentage of intron-containing genes that undergo alternative splicing to 40%. These newly identified transcripts greatly increase the diversity of the maize proteome, sometimes coding for entirely different proteins compared with their most similar annotated isoform. In addition to increasing proteome diversity, many genes encoding novel transcripts gained an additional layer of regulation by microRNAs, often in a tissue-specific manner. We also demonstrate that the majority of genotype-specific alternative splicing can be genetically mapped, with cis -acting quantitative trait loci (QTLs) predominating. A large number of trans -acting QTLs were also apparent, with nearly half located in regions not shown to contain genes associated with splicing. Taken together, these results highlight the currently underappreciated role that alternative splicing plays in tissue identity and genotypic variation in maize. [ABSTRACT FROM AUTHOR]

Abstract: Recently, image representation based on bag-of-visual-words (BoW) model has been popularly applied in image and vision domains. In BoW, a visual codebook of visual words is defined, usually by clustering local features, to represent any novel image with the occurrence of its contained visual words. Given a set of images, we argue that the significance of each image is determined by the significance of its contained visual words. Traditionally, the significances of visual words are defined by term frequency-inverse document frequency (tf-idf), which cannot necessarily capture the intrinsic visual context. In this paper, we propose a new scheme of latent visual context learning (LVCL). The visual context among images and visual words is formulated from latent semantic context and visual link graph analysis. With LVCL, the importance of visual words and images will be distinguished from each other, which will facilitate image level applications, such as image re-ranking and canonical image selection. We validate our approach on text-query based search results returned by Google Image. Experimental results demonstrate the effectiveness and potentials of our LVCL in applications of image re-ranking and canonical image selection, over the state-of-the-art approaches. [Copyright &y& Elsevier]

Sign language recognition (SLR) is a significant and promising technique to facilitate the communication for the hearing-impaired people. In this paper, we are dedicated to weakly supervised continuous SLR, where for each sign video, there are only ordered gloss labels without temporal boundary along frames. To explicitly align video frames to the sign words in a sign video, we propose a novel semantic boundary detection method based on reinforcement learning for accurate continuous SLR. In our approach, we first propose a multi-scale perception scheme to learn discriminative representation for video clips. Then, we formulate the semantic boundary detection as a reinforcement learning problem. We define the state as the feature representation of a video segment, and the action as the determination of the semantic boundary’s location. The reward is computed by the quantitative performance metric between the prediction sentence and the ground truth sentence. The policy network is trained with a policy gradient algorithm. Extensive experiments are conducted on CSL Split II and RWTH-PHOENIX-Weather 2014 datasets, and the results demonstrate the effectiveness and superiority of our method. [ABSTRACT FROM AUTHOR]

The k-nearest neighbors outlier detection is a simple yet effective widely renowned method in data mining. The actual application of this model in the big data domain is not feasible due to time and memory restrictions. Several distributed alternatives based on MapReduce have been proposed to enable this method to handle large-scale data. However, their performance can be further improved with new designs that fit with newly arising technologies. Furthermore, it gives to each attribute the same importance to outlier. There are several approaches to enhance its precision, with the entropy-based outlier detection being among the most successful ones. Entropy-based outlier detection computes attribute entropy of the data set to weighted distance formula for the outlier detection. Apart from the existing the k-nearest neighbors outlier detection to handle big datasets, there is not an entropy-based outlier detection to manage that volume of data. In this paper, we propose an entropy-based outlier detection based on Spark. It presents three separately stages. The first stage computes attribute entropy. The second stage finds the k nearest neighbors and calculates the degrees of outliers using the attribute entropy computed previously. The third stage ranks each point on the degrees of outliers and declares the top n points in this ranking to be outliers. Extensive experimental results show the advantages of the proposed method. This algorithm can improve the outlier detection precision, reduce the runtime and realize the effective large scale dataset outlier detection. [ABSTRACT FROM AUTHOR]

• It is an original work on exploring the region-level diversity and invariance inside images for better learning visual pre-training knowledge. • Propose a simple but effective approach to fully extract the multi-grained visual knowledge inside each image. • Extensive experiments have shown the superiority of our proposed method compared with recent state-of-the-art visual pre-training methods. Recently, self-supervised learning methods have achieved remarkable success in the visual pre-training task. By simply pulling the different augmented views of each image together or other novel mechanisms, they can learn much unsupervised knowledge and significantly improve the transfer performance of pre-training models. However, these works still cannot avoid the representation collapse problem, i.e. , they only focus on limited regions or the extracted features on totally different regions inside each image are nearly the same. Generally, this problem makes the pre-training models cannot sufficiently describe the multi-grained information inside images, which further limits the upper bound of their transfer performance. To alleviate this issue, this paper introduces a simple but effective mechanism, called E xploring the D iversity and I nvariance in Y ourself (E-DIY). By simply pushing the most different regions inside each augmented view away, E-DIY can preserve the diversity of extracted region-level features. By pulling the most similar regions from different augmented views of the same image together, E-DIY can ensure the robustness of extracted region-level features. Benefiting from the above diversity and invariance exploring mechanism, E-DIY better extracts the multi-grained visual information inside each image compared with previous self-supervised learning approaches. Extensive experiments on various downstream tasks have demonstrated the superiority of our method, e.g. , there is 1.9 % improvement (compared with the recent state-of-the-art method, BYOL) on A P 50 metric of detection task on VOC. [ABSTRACT FROM AUTHOR]

Recently, image representation by vector of locally aggregated descriptors (VLADs) has been demonstrated to be super efficient in image representation. Due to the coarse division in the feature space, its discriminative power is limited. One intuitive way to address this issue is to construct a VLAD with a larger vocabulary, but this will lead to a higher dimensional VLAD and suffer more computational complexity when learning the principal component analysis parameters used to project VLAD onto a low-dimensional space. In this paper, we propose a hierarchical scheme to build the VLAD. In our approach, by generating some subwords to each visual word of a coarse vocabulary, a hidden layer visual vocabulary is constructed. With the hidden layer visual vocabulary, the feature space is divided finer. Then, we aggregate the residues in the hidden layer vocabulary to the coarse layer to obtain an image descriptor that is of the same dimension as the original VLAD. In addition, we reveal that performing the whitening operation to local descriptor can further enhance the discriminative power of the VLAD. We validate our approach with experiments mainly conducted on three benchmark data sets, i.e., Holidays data set, UKBench data set, and Oxford Building data set with Flickr1M as distractors and make comparison with the related algorithms on VLAD. The experimental results demonstrate the effectiveness of our algorithm. [ABSTRACT FROM PUBLISHER]

In computer vision, an important challenge to deep neural networks comes from adjusting the varying properties of different image domains. To study this problem, researchers have been investigating a practical setting in which the deep neural networks are trained on a labeled source domain and then transferred to an unlabeled or even unseen target domain. The major difficulty lies in the potential domain gap, which essentially arises from the overfitting in the source domain. Hence, it is important to introduce generalized priors to alleviate the issue. From this perspective, this paper presents a novel framework that forces visual features to align with domain-agnostic priors (DAP). Specifically, we study two kinds of priors, (i) language-guided embedding and (ii) class-level relationship, and we believe that more such priors can be constructed. Our framework, referred to as DAP, is evaluated on both unsupervised domain adaptation (UDA) and domain generalization (DG) where the target domain is unlabeled and even unseen, respectively. We use the standard benchmark that performs transfer semantic segmentation on synthesized datasets (i.e., GTAv and SYNTHIA) and a real dataset (i.e., Cityscapes). Experiments validate the effectiveness of DAP with competitive accuracy in all tasks. In particular, language-guided priors work sufficiently well for UDA, while class-level priors serve as useful complements for DG. The proposed frameworks shed light that domain transfer benefits from better proxies, possibly from other modalities. [ABSTRACT FROM AUTHOR]

In active object tracking, given a visual observation as input, the goal is to lockup the target by autonomously adjusting camera’s position and posture. Previous works on active tracking assume that there is only one object (person) in the environment without distractors. In this work, towards realistic setting, we move forward to a more challenging scenario, where the tracker moves freely in 3D space like unmanned aerial vehicles (UAV) to track a person in various complex scenes with multiple distractors. To this end, we propose a novel end-to-end anti-distractor active object tracking framework by introducing multiple attention modules. On one hand, we take the target template to learn an embedding as channel-wise attention for current observation to distinguish the target from the distractors. On the other hand, temporal attention is introduced to fuse the observation history to extract a feature representation, which is then fed into a reinforcement learning network to output the action of the tracker. To evaluate our method, we build several multi-object 3D environments in Unreal Engine and extensive experiments demonstrate the effectiveness of our approach. [ABSTRACT FROM AUTHOR]

In recent years, there has been growing interest in mapping visual features into compact binary codes for applications on large-scale image collections. Encoding high-dimensional data as compact binary codes reduces the memory cost for storage. Besides, it benefits the computational efficiency since the computation of similarity can be efficiently measured by Hamming distance. In this paper, we propose a novel flexible scale invariant feature transform (SIFT) binarization (FSB) algorithm for large-scale image search. The FSB algorithm explores the magnitude patterns of SIFT descriptor. It is unsupervised and the generated binary codes are demonstrated to be dispreserving. Besides, we propose a new searching strategy to find target features based on the cross-indexing in the binary SIFT space and original SIFT space. We evaluate our approach on two publicly released data sets. The experiments on large-scale partial duplicate image retrieval system demonstrate the effectiveness and efficiency of the proposed algorithm. [ABSTRACT FROM AUTHOR]

Abstract: A novel region-based active contour model (ACM) is proposed in this paper. It is implemented with a special processing named Selective Binary and Gaussian Filtering Regularized Level Set (SBGFRLS) method, which first selectively penalizes the level set function to be binary, and then uses a Gaussian smoothing kernel to regularize it. The advantages of our method are as follows. First, a new region-based signed pressure force (SPF) function is proposed, which can efficiently stop the contours at weak or blurred edges. Second, the exterior and interior boundaries can be automatically detected with the initial contour being anywhere in the image. Third, the proposed ACM with SBGFRLS has the property of selective local or global segmentation. It can segment not only the desired object but also the other objects. Fourth, the level set function can be easily initialized with a binary function, which is more efficient to construct than the widely used signed distance function (SDF). The computational cost for traditional re-initialization can also be reduced. Finally, the proposed algorithm can be efficiently implemented by the simple finite difference scheme. Experiments on synthetic and real images demonstrate the advantages of the proposed method over geodesic active contours (GAC) and Chan–Vese (C–V) active contours in terms of both efficiency and accuracy. [Copyright &y& Elsevier]