Infrared-Visible person RE-IDentification (IV-REID) is a rising task. Compared to conventional re-identification (re-ID), IV-REID concerns the additional modality discrepancy originated from different imaging processes of spectrum cameras, in addition person's appearance caused by viewpoint changes, pose variations and deformations presented re-ID The co-existed discrepancies make more difficult solve. Previous methods attempt reduce simultaneously using feature-level constraints. It however...

Component-based development (CBD) is an important emerging topic in software engineering, promising long-sought-after benefits like increased reuse, reduced time to market, and, hence, production cost. The cornerstone of a CBD technology its underlying component model, which defines components and their composition mechanisms. Current models use objects or architectural units as components. These are not ideal for reuse systematic composition. In this paper, we survey analyze current...

Visible thermal person re-identification (VT-REID) is a task of matching images captured by and visible cameras, which an extremely important issue in night-time surveillance applications. Existing cross-modality recognition works mainly focus on learning sharable feature representations to handle the discrepancies. However, apart from discrepancy caused different camera spectrums, VT-REID also suffers large intra-modality variations environments human poses, so on. In this paper, we propose...

Person reidentification is a key technique to match different persons observed in nonoverlapping camera views. Many researchers treat it as special object-retrieval problem, where ranking optimization plays an important role. Existing methods mainly utilize the similarity relationship between probe and gallery images optimize original list, but seldom consider dissimilarity relationship. In this paper, we propose use both cues framework for person reidentification. Its core idea that true...

Rainy weather is a challenge for many vision-oriented tasks ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e.g.</i> , object detection and segmentation), which causes performance degradation. Image deraining an effective solution to avoid drop of downstream vision tasks. However, most existing methods either fail produce satisfactory restoration results or cost too much computation. In this work, considering both effectiveness efficiency...

Entangled representation of clothing and identity (ID)-intrinsic clues are potentially concomitant in conventional person Re- IDentification (ReID). Nevertheless, eliminating the negative impact on ID remains challenging due to lack theory difficulty isolating exact implications. In this paper, a causality-based Auto-Intervention Model, referred as AIM <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> Codes will publicly available at...

Images captured in low-light or underwater environments are often accompanied by significant degradation, which can negatively impact the quality and performance of downstream tasks. While convolutional neural networks (CNNs) Transformer architectures have made progress computer vision tasks, there few efforts to harmonize them into a more concise framework for enhancing such images. To this end, study proposes aggregate individual capability self-attention (SA) CNNs accurate perturbation...

Person re-identification, aiming to identify images of the same person from various cameras configured in different places, has attracted much attention multimedia retrieval community. In this problem, choosing a proper distance metric is crucial aspect, and many classic methods utilize uniform learnt metric. However, their performance limited due ignoring zero-shot fine-grained characteristics presented real re-identification applications. paper, we investigate two consistencies across...

The performance of traditional face recognition systems is sharply reduced when encountered with a low-resolution (LR) probe image. To obtain much more detailed facial features, some super-resolution (SR) methods have been proposed in the past decade. basic idea image SR to generate high-resolution (HR) from an LR one help set training examples. It aims at transcending limitations optical imaging systems. In this paper, we regard as interpolation problem for domain-specific images. A missing...

Face image super-resolution has attracted much attention in recent years. Many algorithms have been proposed. Among them, sparse representation (SR)-based face approaches are able to achieve competitive performance. However, these SR-based only perform well under the condition that input is noiseless or small noise. When corrupted by large noise, reconstruction weights (or coefficients) of low-resolution (LR) patches using will be seriously unstable, thus leading poor results. To this end,...

Hyperspectral imaging (HSI) classification has become a popular research topic in recent years, and effective feature extraction is an important step before the task. Traditionally, spectral techniques are applied to HSI data cube directly. This paper presents novel algorithm for by exploiting curvelet-transformed domain via relatively new processing technique—singular spectrum analysis (SSA). Although wavelet transform been widely analysis, curvelet employed this since it able separate...

Person re-identification (REID) is an important task in video surveillance and forensics applications. Most of previous approaches are based on a key assumption that all person images have uniform sufficiently high resolutions. Actually, various low-resolutions scale mismatching always exist open world REID. We name this kind problem as Scale-Adaptive Low Resolution Re-identification (SALR-REID). The most intuitive way to address increase (not only low, but also with different scales)...

Existing inpainting methods have achieved promising performance in recovering defective images of specific scenes. However, filling holes involving multiple semantic categories remains challenging due to the obscure se-mantic boundaries and mixture different textures. In this paper, we introduce coherence priors between semantics textures which make it possible concentrate on completing separate a semantic-wise manner. Specifically, adopt multi-scale joint optimization framework first model...

Most person re-identification (ReID) approaches assume that images are captured under relatively similar illumination conditions. In reality, long-term retrieval is common, and often different conditions at times across a day. this situation, the performances of existing ReID models degrade dramatically. This paper addresses problem with variations names it as {\em Illumination-Adaptive Person Re-identification (IA-ReID)}. We propose an Illumination-Identity Disentanglement (IID) network to...

Visible-infrared person re-identification (RGB-IR ReID) is extremely important for the surveillance applications under poor illumination conditions. Since difference in feature representations not only lies person' pose, viewpoint or variations, but also comes from huge spectrum discrepancy, task becomes practically very challenging. Existing RGB-IR ReID models focus on bridging gap between RGB and IR images through shared embedding, subspace learning via adversarial learning. However, these...

Person re-identification (Re-ID) aims at matching person images captured in non-overlapping camera views. To represent appearance, low-level visual features are sensitive to environmental changes, while high-level semantic attributes, such as "short-hair" or "long-hair", relatively stable. Hence, researches have started design attributes reduce the ambiguity. However, train a prediction model for it requires plenty of annotations, which hard obtain practical large-scale applications....

Convolutional neural network (CNN) and Transformer have achieved great success in multimedia applications. However, little effort has been made to effectively efficiently harmonize these two architectures satisfy image deraining. This paper aims unify take advantage of their learning merits for In particular, the local connectivity translation equivariance CNN global aggregation ability self-attention (SA) are fully exploited specific context structure representations. Based on observation...

Low-light image enhancement aims to improve an image's visibility while keeping its visual naturalness. Different from existing methods, which tend accomplish the relighting task directly, we investigate intrinsic degradation and relight low-light refining details color in two steps. Inspired by formulation (diffuse illumination plus environment color), first estimate inputs simulate distortion of color, then refine content recover loss diffuse color. To this end, propose a novel...

Recent person Re-IDentification (ReID) systems have been challenged by changes in personnel clothing, leading to the study of Cloth-Changing ReID (CC-ReID). Commonly used techniques involve incorporating auxiliary information ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">e.g</i> ., body masks, gait, skeleton, and keypoints) accurately identify target pedestrian. However, effectiveness these methods heavily relies on quality comes at cost...

Adversarial attacks on thermal infrared imaging expose the risk of related applications. Estimating security these systems is essential for safely deploying them in real world. In many cases, realizing physical space requires elaborate special perturbations. These solutions are often impractical and attention-grabbing. To address need a physically practical stealthy adversarial attack, we introduce HotCold Block, novel attack detectors that hide persons utilizing wearable Warming Paste...

Spatial-Temporal Video Super-Resolution (ST-VSR) aims to generate high-quality videos with higher resolution (HR) and frame rate (HFR). Quite intuitively, pioneering two-stage based methods complete ST-VSR by directly combining two sub-tasks: Spatial (S-VSR) Temporal (T-VSR) but ignore the reciprocal relations among them. 1) T-VSR S-VSR: temporal correlations help accurate spatial detail representation; 2) S-VSR T-VSR: abundant information contributes refinement of prediction. To this end,...

Current visible-infrared cross-modality person re-identification research has only focused on exploring the bi-modality mutual retrieval paradigm, and we propose a new more practical mix-modality paradigm. Existing V isible- I nfrared (VI-ReID) methods have achieved some results in paradigm by learning correspondence between visible infrared modalities. However, significant performance degradation occurs due to modality confusion problem when these are applied Therefore, this paper proposes...

Person reidentification (re-id), as an important task in video surveillance and forensics applications, has been widely studied. Previous research efforts toward solving the person re-id problem have primarily focused on constructing robust vector description by exploiting appearance's characteristic, or learning discriminative distance metric labeled vectors. Based cognition identification process of human, we propose a new pattern, which transforms feature from characteristic to...