Search

• An adaptive HEVC video steganography algorithm is designed, which can effectively resist the PU shift-based steganalysis. • The introduction of multivariate STC allows for smaller modifications with the same embedding capacity. • The superiority of this algorithm over the state-of-the-art video steganography algorithm is verified through extensive experiments. Calibration-based steganalysis methods are a fatal threat to current HEVC steganography algorithms. The current steganography algorithms based on prediction unit (PU) still lack resistance to this type of steganalysis. The reason is that the improvements in HEVC codec provide more potential steganographic space as well as introduce more detectable distortion. Therefore, a HEVC video steganography method that has resistance to PU shift-based steganalysis is proposed in this paper. First, the PU shift phenomenon that exists in PU-based steganography is introduced. Then the existing calibration-based steganalysis method is enhanced. According to the encoding rules of HEVC, the distortion cost function according to the difference of rate distortion cost (RD cost) during recompression is designed, which aims to obtain better resistance to calibration-based steganalysis. Experimental results demonstrate that the steganography algorithm designed in this paper provides excellent resistance ability for calibration-based steganalysis. [ABSTRACT FROM AUTHOR]

The correlation filter based online tracking methods usually can achieve high real-time performance due to the leverage of the well-known FFT. However, they are also apt to generate the "corrupted training samples" in scenarios with complex background, which will trigger the model drift and deteriorate the tracking accuracy rapidly. The existing methods usually consider this problem from certain aspect and none of them has mined the potential of combining multiple formulas. In this paper, we propose the Output Constraint Transfer - Discriminative Scale Space Tracking (OCT-DSST) algorithm, which has taken full consideration of multiple channel feature, multiple filters, kernel trick, memory with incremental learning, and the self-supervision mechanism. We re-formulate the online tracking process by combining all formulas above in a unified framework. The so obtained adaptive learning rate can better exploit the feedback information coming from the intermediate tracking results, and effectively mitigate the corrupted sample problem. The experimental results on the OTB-50/100 and the VOT2016 datasets reveal that the proposed method is comparative to most state-of-the-arts algorithms, and can increase the accuracy by 2% and the success rate by 1.7%, compared to the traditional DSST method. [ABSTRACT FROM AUTHOR]

• An original classification of attacks on robust watermarks is proposed. • A review of watermarking schemes designed to counter targeted attacks is presented. • The main methods of countering targeted attacks are noted. • Promising research directions are discussed. Image watermarking is an effective and promising technology. Robust watermarks that are resistant to various attacks allow authors and owners of digital images to protect their rights to digital content, control its distribution and confirm its authenticity. Most of the modern algorithms for robust image watermarking aim to achieve resistance to a large number of different attacks. However, some authors develop algorithms designed to counter targeted attacks. The study of such schemes allows developers of watermarking algorithms to evaluate special means of counteracting various attacks, and then use them to create new robust schemes, both targeted and universal ones. In this paper, we present an overview of robust image watermarking schemes in terms of countering targeted attacks. We review the state-of-the-art in the field of attacking robust watermarks and propose a four-level classification of attacks that includes different levels of attack implementation, including an attacker's intent, characteristics of actions, the main target and an attack type. The proposed classification considers a watermark as an object of attack and summarizes various characteristics of attacks in a hierarchical manner. We analyze the means of countering common attacks such as image processing attacks, geometric attacks, print-scan and screen capture attacks, collusion attacks, and ambiguity attacks. Based on the results of our review, we highlight the most common methods of countering attacks and formulate promising areas of research in the field of methods for improving security of embedding schemes. [ABSTRACT FROM AUTHOR]

Researchers have been involved for decades in search of an efficient skin detection method. However, current methods have not overcome the significant challenges of skin detection, such as variation of illumination, various skin tones of different ethnic groups, and many others. This research proposed a clustering and region-growing-based skin detection method to overcome these limitations. Together with significant insight, these methods result in a more effective algorithm. The insight concerns the capability to dynamically define the number of clusters in a collection of pixels organized as images. In Clustering for most problem domains, the number of clusters is fixed prior and does not perform effectively over a wide variety of data contents. Therefore, this research paper proposed a skin detection method that validated the above findings. The proposed method assigns the number of clusters based on image properties and ultimately allows freedom from manual thresholds or other manual operations. The dynamic determination of clustering outcomes allows for greater automation of skin detection when dealing with uncertain real-world conditions. • Human skin tone is uniform, and clustering is an efficient method to detect human skin. • CIELAB color space is better for separating illumination from the image. • A small number of significant objects can occupy most of the pixel in an image. • Pixel connectivity of an image can automatically decide the number of clusters. • Clustering performs efficiently in Human skin detection. [ABSTRACT FROM AUTHOR]

Domain Generalization (DG) model is an important tool to improve the robustness of person re-identification algorithm, but the domain gap makes it difficult to transfer knowledge cross-domain effectively. To solve the above problems, this paper proposes a generalization model based on a Meta-Bond Domain Alignment (M−BDA) model. To learn a generalizable model, a meta -learning strategy is introduced to simulate the training–testing process of the domain generalization. Then a bond-domain module is constructed in the training to align the source domain, which can reduce the domain gap between the source domain and the target domain, and facilitate the knowledge transfer. Finally, the bond-domain loss is counted on the feature space of the bond-domain to prevent the generated bond-domain from overfitting to the source domain. The experimental results show that the proposed algorithm achieves better performance in DukeMTMC-ReID, Market1501 and MSMT17 alternating as the source and the target domain tasks. In the generalization experiments of Market-1501 → DukeMTMC-ReID, mAP and Rank-1 increased by 7.7 % and 3.6 %, respectively, and in the generalization experiments of DukeMTMC-ReID → Market-1501, mAP and Rank-1 are increased by 9.3 % and 3.5 %, respectively, which are significantly better than the newer representative algorithms. [ABSTRACT FROM AUTHOR]

Locally Sensitive Hashing (LSH) algorithms are classical algorithms commonly used on the c-Approximate Nearest Neighbor (c-ANN) search problem. When using Euclidean distance to measure sample similarity and solve the c-ANN problem, the traditional approach is to utilize the Exact Euclidean Locality Sensitive Hashing (E2LSH) algorithm based on the p-stable distribution. However, the uncertainty of the p-stable distribution causes the hash buckets constructed by the E2LSH algorithm to vary in queries. Therefore, this paper proposes the OLLSH algorithm based on the Weighted Majority algorithm in the Online-Learning framework, which selects the hash buckets with more stable query accuracy by weighted voting on the hash buckets generated by the E2LSH algorithm. Then, we conduct simulation experiments on synthetic dataset and four real data sets and conclude that the proposed OLLSH algorithm improves the accuracy compared to the original algorithm with the same memory usage. • The parameters of the E2LSH algorithm are analysed in detail. • The choice of hash buckets for the E2LSH algorithm is achieved with the WM algorithm. • The OLLSH algorithm saves memory by using less hash buckets during the query process. • The efficacy of the OLLSH algorithm was proved on both synthetic and real datasets. [ABSTRACT FROM AUTHOR]

Existing stereo matching algorithms suffer from issues such as susceptibility to distortion, weak noise resistance, and a high rate of mismatches in regions with weak textures and discontinuous disparities. To address these challenges, this paper proposes a stereo matching algorithm based on an improved census transform and minimum spanning tree (MST) cost aggregation. In the cost calculation phase, we employ a Gaussian-weighted transformation window and incorporate gradient and edge information to perform weighted fusion of the results. In the cost aggregation process, we introduce a collaborative adaptive window. Each pixel acquires information from the support window of the guided filter (GF) and other pixels within the MST. Furthermore, we integrate the SLIC superpixel segmentation algorithm into MST construction. These two components work synergistically to assign appropriate adaptive weights to pixels, facilitating coordinated cost volume aggregation. Different optimization methods are applied to address mismatched points of various types in post-disparity processing.Performance evaluation using the Middlebury dataset and KITTI dataset demonstrates that our proposed algorithm not only enhances matching accuracy in regions with discontinuous disparities and weak textures but also exhibits significantly improved robustness to interference. Additionally, the resulting disparity map displays smoother edges. • Cost Calculation Phase: Introducing a Gaussian-weighted transformation window, along with gradient and edge information, enables weighted fusion of results. • Cost Aggregation Process: Collaborative adaptive window introduction allows each pixel to gather information from the support window of the guided filter and neighboring pixels within the Minimum Spanning Tree(MST).Furthermore,the SLIC superpixel segmentation algorithm is integrated into the construction of the MST. • Post-Disparity Processing Optimization: Applying diverse optimization methods addresses mismatches among different types of points. [ABSTRACT FROM AUTHOR]

• A small target detection algorithm based on feature interaction and guided learning. • A feature guided alignment module is designed to obtain more effective features. • Gaussian distribution sample allocation strategy is introduced to obtain better samples. • Feature interaction and decoupling detection enhance the detection task capability. • A new regression loss function is used to optimize the localization ability to a certain extent. At present, for the diverse target sizes, different angles, and complex environment in aerial images, how to detect small objects effectively in aerial images is still a challenge task. In order to further improve the detection accuracy and reduce the missing rate, in this paper, we proposes a small target detection algorithm based on feature interaction and guided learning. Firstly, the feature guided alignment module is designed to deal with the problem of information loss and overlap. Secondly, the misallocation of small object samples is optimized by introducing gaussian sample allocation strategy. Then the interactive parallel detection header is designed to solve the detection task conflict problem. Finally, the new regression loss function is designed to enhance the localization ability of the target. The experimental results on the Tinyperson dataset and Visdrone dataset show that this model can improve the detection accuracy and reduce the missing rate of small targets effectively. [ABSTRACT FROM AUTHOR]

Object drift determination is a crucial issue in long-term tracking. Most existing object drift determination criteria require manually selecting different thresholds on different datasets to determine whether the object is lost. In this case, choosing the appropriate threshold is a complex problem. An object drift determination network based on dual-template joint decision-making is proposed to address this issue. The proposed object drift determination network not only does not require selection of thresholds on different datasets, and can be used as a plug-and-play module of short-term visual tracking algorithm to achieve long-term visual tracking tasks with good generalization ability. The proposed object drift determination network is applied to four short-term baseline trackers and constructs four long-term visual tracking algorithms. Experimental results verify that all four improved algorithms significantly improve long-term visual tracking performance compared to the original algorithms. In addition, the determined speed of the object drift determination network proposed in this paper reaches 111 FPS, which has little effect on the long-term visual tracking speed. • New drift determination network without threshold selection is proposed. • New template update criterion is proposed to ensure the quality of templates. • Capable of acting as a plug-and-play module for most basic trackers. • The effectiveness of the proposed method is verified on five public datasets. [ABSTRACT FROM AUTHOR]

• A layout plane is a closed loop surrounded by layout keypoints. • The wall structure can be parametrization as an undirected graph. • Layout keypoint prediction is relative to human pose estimation. • Cuboid room assumption is limited in real life. Room layout estimation aims to predict the spatial structure of a room from a single image. Most existing methods relying on 2D cues are suitable for cuboid rooms, but non-cuboid rooms. 3D layout estimation methods can reconstruct 3D models of general rooms, but they are trained with depth information on high collection costs, consume large computation resources, and run slowly. This paper considers an undirected graph representation method of the general room layouts, which includes cuboid and non-cuboid rooms consisting of a single ceiling, a single floor, and multiple walls. To this end, we first predict the positions of the layout vertices and then use the network automatically learn the connection relationship between the vertices. The final layout is obtained through a simple layout inference post-processing algorithm. The experimental results both on cuboid and non-cuboid datasets validate the effectiveness and efficiency of our method. The code is available at https://github.com/Hui-Yao/2D-graph-layout-estimation. [ABSTRACT FROM AUTHOR]

To overcome challenges in multiple-object tracking (MOT) tasks, recent algorithms use interaction cues alongside motion and appearance features. These algorithms use graph neural networks or transformers to extract interaction features that lead to high computation costs. In this paper, a novel interaction cue based on geometric features is presented aiming to detect occlusion and reidentify lost targets with low computational costs. Moreover, in the majority of algorithms, camera motion is considered negligible, which is a strong assumption that is not always true and can lead to identity (ID) switching or mismatching of targets. In this paper, a method for measuring camera motion is presented that efficiently reduces its effect on tracking. The proposed algorithm is evaluated on MOT17 and MOT20 datasets and achieves state-of-the-art performance on MOT17 with comparable results on MOT20. The code is also publicly available. 1 1 https://github.com/mhnasseri/for_tracking. • Cascade association based on the detection score. • A new similarity metric that considers detection's size. • A novel interaction model for occlusion handling and target re-identification. • A novel algorithm for estimating camera motion and removing it. • State-of-the-art results on MOT17 and comparable results on MOT20. [ABSTRACT FROM AUTHOR]

The latest Versatile Video Coding (VVC) standard incorporates a series of effective and complex new intra coding tools, which obtains superior coding efficiency than the High Efficiency Video Coding (HEVC). However, this makes the intra coding more complicated and time-consuming. A fast algorithm for VVC from two aspects of fast mode decision and fast partition decision is proposed in this paper. For the fast mode decision, the relationship between bins with Histogram of Oriented Gradient (HOG) and intra modes is created for the mode selection, decreasing the planar modes for SATD and RDO. Moreover, we analyze the maximum bins to determine the final modes, and we use the modes of left and upper blocks as a reference for the current CU, which can early terminate RDO. Moreover, a two-step fast partition algorithm is proposed based on HOG for fast partition decision, in which two thresholds are investigated to control the uniformity of textures. The proposed fast algorithm is implemented on the VVC test model, and the experimental results show that it can achieve 69.07% time savings with only 2.96% BDBR increases averagely, which outperforms other relatively existing state-of-the-art methods. Moreover, to convince the universality of our algorithm, we further implement our method in Fraunhofer Versatile Video Encoder (VVenc) and Fraunhofer Versatile Video Decoder (VVdec), which have five settings to control the trade-off between encoding quality and efficiency for intra coding. The fast intra mode decision algorithm and fast partition algorithm decrease the complexity of intra coding for both VTM and VVenc, which shows the efficiency and universality of the proposed fast partition and fast mode decision algorithms. • The CUs' textures have strong correlations with the HOG operator. • A mapping table of VVC intra modes with HOG operator is created. • The HOG operator leads the intra mode decision to decline the probability and number of RDO operations. • The HOG operator controls the trade-off between BDBR and saving time with two VVC intra partition decision thresholds. • The VVC intra coding is accelerated with fast intra mode and fast partition decisions with HOG operator. [ABSTRACT FROM AUTHOR]

Most deep learning-based image enhancement algorithms have been developed based on the image-to-image translation approach, in which enhancement processes are difficult to interpret. In this paper, we propose a novel interpretable image enhancement algorithm that estimates multiple transformation functions to describe complex color mapping. First, we develop a histogram-based multiple transformation function estimation network (HMTF-Net) to estimate multiple transformation functions by exploiting both the spatial and statistical information of the input images. Second, we estimate pixel-wise weight maps, which indicate the contribution of each transformation function at each pixel, based on the local structures of the input image and the transformed images obtained by each transformation function. Finally, we obtain the enhanced image as the weighted sum of the transformed images using the estimated weight maps. Extensive experiments confirm the effectiveness of the proposed approach and demonstrate that the proposed algorithm outperforms state-of-the-art image enhancement algorithms for different image enhancement tasks. [ABSTRACT FROM AUTHOR]

• Rethinking the PASCAL-VOC and MS-COCO dataset for small object detection. • There are different labeling errors in these two benchmarks by visual analysis. • Building several specific datasets to analyze small object detection. • Data quality is another factor that affects the performance of small object detection. The data and the algorithm are critical to deep learning-based small object detectors. In this paper, we rethink the PASCAL-VOC and MS-COCO dataset for small object detection. By visual analysis of the original annotations, we find that there are different labeling errors in these two datasets. To solve these problems, we build specific datasets, including SDOD, Mini6K, Mini2022 and Mini6KClean. The experimental results of several typical algorithms (e.g. SSD, YOLOv5, Faster RCNN and Deformable DETR) on the datasets show that data labeling errors (such as missing labels, category label errors, inappropriate labels) are another factor that affects the detection performance of small objects. [ABSTRACT FROM AUTHOR]

• Build a correlation model of adjacent features to guide the feature update of subsequent states. • The updated features are fused with the template features to filter the discriminative attribute features. • Simulate the saccade mechanism to solve the problem of missing fine-grained features in batch information processing. • Filter fine-grained features by using hierarchical guidance and multiple traversals. The tracker is a core component of the tracking algorithm, but it is difficult to identify the object, which is a challenge to improve the tracking accuracy. This paper proposes a Siamese network-based tracking algorithm based on tensor space mapping and memory-learning mechanisms. Firstly, the source image is mapped to the tensor space to serialize the feature distributions. Then the gating mechanism is used to extract the association information about the adjacent state, which guides the update of the subsequent state, and the interactive information on the objects is used to locate the object. On this basis, a memory-learning module is built to traverse and extract the fine-grained features, which can filter the semantic information of the object learned by the tracker. As a result, the tracking accuracy is enhanced. The experiments show that the proposed algorithm has better performance than that of the comparison methods in the OTB100 data set and the VOT data set. [ABSTRACT FROM AUTHOR]

Schemes to complement context relationships by cross-scale feature fusion have appeared in many RGB-D scene parsing algorithms; however, most of these works conduct multi-scale information interaction after multi-modal feature fusion, which ignores the information loss of the two modes in the original coding. Therefore, a cross-complementary fusion network (CCFNet) is designed in this paper to calibrate the multi-modal information before feature fusion, so as to improve the feature quality of each mode and the information complementarity ability of RGB and the depth map. First, we divided the features into low, middle, and high levels, among which the low-level features contain the global details of the image and the main learning features include texture, edge, and other features. The middle layer features contain not only some global detail features but also some local semantic features. Additionally, the high-level features contain rich local semantic features. Then, the feature information lost in the coding process of low and middle level features is supplemented and extracted through the designed cross feature enhancement module, and the high-level features are extracted through the feature enhancement module. In addition, the cross-modal fusion module is designed to integrate multi-modal features of different levels. The experimental results verify that the proposed CCFNet achieves excellent performance on the RGB-D scene parsing dataset containing clothing images, and the generalization ability of the model is verified by the dataset NYU Depth V2. [ABSTRACT FROM AUTHOR]

In current society, artificial intelligence processing technology offers convenient video monitoring, but also raises the risk of privacy leakage. Theoretically, the data used in intelligent video processing methods may directly convey visual information containing private content. For the above problem, this paper uses a multi-layer visual privacy-protected (VPP) coding method to blur private content in the video at the visual level, while avoiding the loss of important visual features contained in the video as much as possible. And this provides a guarantee of the quality of the subsequent keyframe extraction step. Then a visual evaluation algorithm is proposed for assessing the quality of VPP-encoded video privacy protection. And the experiment shows that the results are consistent with those of subjective evaluation. In addition, for VPP-encoded video, we propose an unsupervised two-layer clustering keyframe extraction method with corresponding performance evaluation index. Finally, an association model is established to balance the privacy protection quality and the keyframe extraction performance. [ABSTRACT FROM AUTHOR]

Fast image codecs are a current need in applications that deal with large amounts of images. Graphics Processing Units (GPUs) are suitable processors to speed up most kinds of algorithms, especially when they allow fine-grain parallelism. Bitplane Coding with Parallel Coefficient processing (BPC-PaCo) is a recently proposed algorithm for the core stage of wavelet-based image codecs tailored for the highly parallel architectures of GPUs. This algorithm provides complexity scalability to allow faster execution at the expense of coding efficiency. Its main drawback is that the speedup and loss in image quality is controlled only roughly, resulting in visible distortion at low and medium rates. This paper addresses this issue by integrating techniques of visually lossless coding into BPC-PaCo. The resulting method minimizes the visual distortion introduced in the compressed file, obtaining higher-quality images to a human observer. Experimental results also indicate 12% speedups with respect to BPC-PaCo. [ABSTRACT FROM AUTHOR]

Cover selection is one of the important techniques to improve the security of image steganography. However, existing methods mostly focus on cover selection of grayscale image. In this paper, we propose a novel Cover Selection Algorithm of color images based on Hybrid Local Texture Descriptor, named HLTD-CSA. A green-channel related Local Binary Pattern (LBP) is designed which utilizes local statistics of intra-channel and cross-channel correlations efficiently. Besides, Local Phase Quantization is introduced to serve as a complementary component of our improved LBP. To further enhance the performance of cover selection for color image, a hybrid local texture descriptor (HLTD) is obtained by combining these two types of local texture descriptors with a proper combination strategy. Finally, the proposed algorithm selects images which have larger values of HLTD to construct the cover image set. Extensive experiments are conducted to verify the effectiveness of our method. • A cover selection method of color image is proposed using hybrid texture descriptor. • A novel local binary pattern is designed with intra- and cross-channel correlations. • Descriptors in spatial and frequency domains are combined with a proper strategy. • Experiments verify the efficiency of improving steganography security by our method. [ABSTRACT FROM AUTHOR]

Nowadays, stereoscopic image quality assessment (SIQA) based on convolutional neural network (CNN) has become the mainstream model of image quality assessment (IQA). Compared with the two-dimensional quality evaluation model, stereoscopic image quality evaluation is more challenging due to the effects of depth and parallax information. In this paper, we propose a two-stream interactive network model to perform quality evaluation, which can well simulate the process of human stereo visual perception. Meanwhile, we enhance the extraction of local and global features of images by asymmetric convolution kernel and interactive sub-networks of inter-layers, respectively, which can further optimize our network model. Our proposed algorithm was evaluated on four public databases. The final experimental results show that our proposed algorithm exhibits good performance not only on the whole database but also on each single distortion type. [ABSTRACT FROM AUTHOR]

Counting objects is a fundamental but challenging problem. In this paper, we propose diffusion-based, geometry-free, and learning-free methodologies to count the number of objects in images. The main idea is to represent each object by a unique index value regardless of its intensity or size, and to simply count the number of index values. First, we place different vectors, refer to as seed vectors, uniformly throughout the mask image. The mask image has boundary information of the objects to be counted. Secondly, the seeds are diffused using an edge-weighted harmonic variational optimization model within each object. We propose an efficient algorithm based on an operator splitting approach and alternating direction minimization method, and theoretical analysis of this algorithm is given. An optimal solution of the model is obtained when the distributed seeds are completely diffused such that there is a unique intensity within each object, which we refer to as an index. For computational efficiency, we stop the diffusion process before a full convergence, and propose to cluster these diffused index values. We refer to this approach as Counting Objects by Diffused Index (CODI). We explore scalar and multi-dimensional seed vectors. For Scalar seeds, we use Gaussian fitting in histogram to count, while for vector seeds, we exploit a high-dimensional clustering method for the final step of counting via clustering. The proposed method is flexible even if the boundary of the object is not clear nor fully enclosed. We present counting results in various applications such as biological cells, agriculture, concert crowd, and transportation. Some comparisons with existing methods are presented. [ABSTRACT FROM AUTHOR]

The more advanced multi-view extension, MV-HEVC, effectively exploits visual similarities between multi-view videos and enables high compression efficiency. Each view in the multi-view sequence depends on the captured scene, the distance between cameras and recording angles. Increasing the distance between dependent viewpoints generates an inter-view disparity. This impacts the inter-view similarities, affects the disparity estimation and further increases the computational complexity of the MV-HEVC encoder. In this paper, an efficient earlier disparity estimation is proposed for low complexity MV-HEVC. This algorithm is based on reducing the complexity of disparity estimation by eliminating the inter-view offset. Moreover, the inter-view similarities are controlled by considering the reliability of each coding unit size in the search range. This reliability is estimated by reducing the number of searching points within a new limited window. For reliable motion estimation, we further proposed an earlier decision of coding units splitting in the dependent views according to those in the reference views. Experimental results show that the proposed algorithm can achieve an average encoding time saving of 20.37%–40,61% with marginal performance degradation. • Visual adjustment between views can be improved by eliminating inter-view offsets. • Eliminating inter-view offsets reduces search time in the disparity estimation. • The visual adjustment allows a trade-off between encoding time and video quality. • An earlier decision of coding units splitting for reliable motion estimation. [ABSTRACT FROM AUTHOR]

Person re-identification(ReID) is an intelligent video surveillance technology that retrieves the same person from different cameras. This task is extremely challenging due to changes in person poses, different camera views, and occlusion. In recent years, person ReID based on deep learning technology has received widespread attention due to the rapid development and excellent performance of deep learning. In this paper, we first divide person ReID based on deep learning approaches into seven types, i.e., fused hand-crafted features deep model, representation learning model, metric learning model, part-based deep model, video-based model, GAN-based model, unsupervised model. Furthermore, we launched a brief overview of the seven types. Then, we introduce some examples of commonly used datasets, compare the performance of some algorithms on image and video datasets in recent years, and analyze the advantages and disadvantages of various methods. Finally, we summarize the possible future research directions of person ReID technology. • Dividing person ReID based on deep learning approaches into seven types. • Introducing some examples of commonly used datasets. • Comparing the performance of some algorithms on image and video datasets in recent years. • Analyzing the advantages and disadvantages of various methods. • Summarizing the possible future research directions of person ReID technology. [ABSTRACT FROM AUTHOR]

• Proposing a model for blind image deconvolution which adopts the L 0 -norm and total variation to regularize the latent image and blur kernel, respectively. • Proposing an automatic algorithm which can simultaneously estimate the intermediate sharp image, blur kernel and regularization parameters. • Proposing a model for non-blind image deconvolution which takes the mixture of L 0 -norm of image gradients and total variation for regularization. • Proposing an algorithm which is easy to be extended to solve the non-blind image deconvolution problem. This paper proposes a blind image deconvolution method which consists of two sequential phases, i.e., blur kernel estimation and image restoration. In the first phase, we adopt the L 0 -norm of image gradients and total variation (TV) to regularize the latent image and blur kernel, respectively. Then we design an alternating optimization algorithm which jointly incorporates the estimation of intermediately restored image, blur kernel and regularization parameters into account. In the second phase, we propose to take the mixture of L 0 -norm of image gradients and TV to regularize the latent image, and design an efficient non-blind deconvolution algorithm to achieve the restored image. Experimental results on both a benchmark image dataset and real-world blurred images show that the proposed method can effectively restore image details while suppress noise and ringing artifacts, the result is of high quality which is competitive with some state of the art methods. [ABSTRACT FROM AUTHOR]

• A sign language recognition framework based on global-local attention, including: • A Multiple Group Random Segment Sampling algorithm to extract the key frame clips; • A global Attention-Model based ResC3D network with an improved timing conversion layer; • A local SLR algorithm based on Mask RCNN hand detection network. Video-level sign language recognition is still a challenging task due to the influence of sign language-independent factors and timing requirements. This paper constructs a sign language recognition framework based on global-local feature description, and proposes a three-dimensional residual global network model with attention layer and a local network model based on target detection. The global feature description is based on the whole video behavior for time series modeling. The improved timing conversion layer is used to explore the timing information of different periods and learn the video representations of different timings. In the local module the hand is located through the target detection network to highlight its key role in the whole sign language behavior, which strengthens the category differences, and compensates the global network. Experiments on two well-known Chinese sign language datasets (SLR_Dataset and DEVSIGN_D) show that the proposed method can obtain higher recognition accuracy (respectively 89.2%, 91%) and better generalization performance. [ABSTRACT FROM AUTHOR]

• We develop a detection algorithm and analyze results for different occluded sizes. • We selectively drop objects' features to enrich the occluded scenes. • We learn the implicit mapping to recover occlude noise to real feature. • We verify the effect of GC-FRCN and apply it to the in-filed Rice Panicles Counting. Detecting the object with external occlusion has always been a hot topic in computer version, while its accuracy is always limited due to the loss of original object information and increase of new occlusion noise. In this paper, we propose a occluded object detection algorithm named GC-FRCN (Generative feature completing Faster RCNN), which consists of the OSGM (Occlusion Sample Generation Module) and OSIM (Occlusion Sample Inpainting Module). Specifically, the OSGM mines and discards the feature points with high category response on the feature map to enhance the richness of occlusion scenes in the training data set. OSIM learns an implicit mapping relationship from occluded feature map to real feature map adversarially, which aims at improving feature quality by repair the noisy object feature. Extensive experiments and ablation studies have been conducted on four different datasets. All the experiments demonstrate the GC-FRCN can effectively detect objects with local external occlusion and has good robustness for occlusion at different scales. [ABSTRACT FROM AUTHOR]