Text-based person search is an important sub-task in cross-modality image retrieval, aiming to capture interested person images by giving textual descriptions. The huge information differences between image and text modalities make this task challenging. Recent methods take local-aligned feature learning strategy into consideration, but lack sufficient mining of more local information. Accordingly, we explore a Multi-level Part-aware Feature Disentangling (MPFD) framework to more fully extract visual and textual representations from multiple angles. Specifically, we introduce a Textual Part-aware Matching (TPM) module into the existing baseline, to disentangle local features for detailed information from both visual and textual part-aware aspects. Besides, in order to fuse multiple local features and improve discrimination of global features, we propose a Multi-level Feature Integration (MFI) module which is capable to perceive the relations between features. We carry out adequate experiments on CUHK-PEDES and ICFG-PEDES datasets to verify our proposed framework, and the results demonstrate that MPFD framework performs favorably against the state-of-the-art methods.

Person re-identification (re-ID) aims to match the same person across different cameras. However, most existing re-ID methods assume that people wear the same clothes in different views, which limit their performance in identifying target pedestrians who change clothes. Cloth-changing re-ID is a quite challenging problem as clothes occupying a large number of pixels in an image becomes invalid or even misleads information. To tackle this problem, we propose a novel Multi-biometric Unified Network (MBUNet) for learning the robustness of cloth-changing re-ID model by exploiting clothing-independent cues. Specifically, we first introduce a multi-biological feature branch to extract a variety of biological features, such as the head, neck, and shoulders to resist cloth-changing. Then, a differential feature attention module (DFAM) is embedded in this branch, which can extract discriminative fine-grained biological features. Besides, we design a differential recombination on max pooling (DRMP) strategy and simultaneously apply a direction-adaptive graph convolutional layer to mine more robust global and pose features. Finally, we propose a Lightweight Domain Adaptation Module (LDAM) that combines the attention mechanism before and after the waveblock to capture and enhance transferable features across scenarios. To further improve the performance of the model, we also integrate mAP optimization into the objective function of our model for joint training to solve the discrete optimization problem of mAP. Extensive experiments on five cloth-changing re-ID datasets demonstrate the advantages of our proposed MBUNet. The code is available at https://github.com/liyeabc/MBUNet.

Person re-identification (re-ID) aims at matching the same person across different cameras. Most of the existing meth-ods for re- ID assume that people wear the same clothes on different cameras. However, Cloth-Changing re- ID is a quite challenging problem since people are likely to change clothes as the time span increases. To tackle this problem, a Multi-Biometric Unified Network (MBUNet) is proposed to ex-ploit clothing-unrelated cues. We firstly introduce a multi-biological feature branch that aims at extracting a variety of biological features, such as the head, neck, and shoulders to resist clothing changes. To extract discriminative fine-grained biological features, we embed a differential feature attention module (DFAM) for it. Besides, we adopt differ-ential recombination on max pooling (DRMP) and apply a direction-adaptive graph convolutional layer to extract more robust global features and pose features. Extensive experi-ments on three Cloth-Changing re-ID datasets show the ad-vantages of our proposed MBUNet.

The performance of person re-identification (re-ID) is easily affected by illumination variations caused by different shooting times, places and cameras. Existing illumination-adaptive methods usually require annotating cross-camera pedestrians on each illumination scale, which is unaffordable for a long-term person retrieval system. The cross-illumination person retrieval problem presents a great challenge for accurate person matching. In this paper, we propose a novel method to tackle this task, which only needs to annotate pedestrians on one illumination scale. Specifically, (i) we propose a novel Illumination Estimation and Restoring framework (IER) to estimate the illumination scale of testing images taken at different illumination conditions and restore them to the illumination scale of training images, such that the disparities between training images with uniform illumination and testing images with varying illuminations are reduced. IER achieves promising results on illumination-adaptive dataset and proving itself a proper baseline for cross-illumination person re-ID. (ii) we propose a Mixed Training strategy using both Original and Reconstructed images (MTOR) to further improve model performance. We generate reconstructed images that are consistent with the original training images in content but more similar to the restored images in style. The reconstructed images are combined with the original training images for supervised training to further reduce the domain gap between original training images and restored testing images. To verify the effectiveness of our method, some simulated illumination-adaptive datasets are constructed with various illumination conditions. Extensive experimental results on the simulated datasets validate the effectiveness of the proposed method. The source code is available athttps://github.com/FadeOrigin/IUReId.

The goal of unsupervised person re-identification (Re-ID) is to use unlabeled person images to learn discriminative features. In recent years, many approaches have adopted clustered pseudo labels to construct proxies for contrastive learning, and have thereby achieved great success. However, existing methods of this kind only utilize local structures within IDs to design their proxies while ignoring the relations between samples of different IDs, which limits the improvement for inter-ID discriminative ability. To resolve this issue, we propose a Global Relation-Aware Contrast Learning (GRACL) method for the task of unsupervised Re-ID. Our method first sets up two proxies for each cluster to capture the inter- and intra-ID relations respectively, which enables us to both effectively increase inter-ID variances and reduce the intra-ID discrepancies. Specifically, the samples that are most different from those in different clusters are selected as inter-ID relation-aware proxies, while those that are least similar to samples from the same clusters are employed as intra-ID relation-aware proxies. With the aid of these proxies, we design both inter- and intra-ID relation-aware contrastive learning modules to facilitate model learning. By pulling each sample close to the positive proxy, we can obtain identity-invariant discriminative features. Experiments on five widely-used Re-ID datasets prove that our GRACL model outperforms current state-of-the-art approaches to a remarkable extent.

Recently, video-based person re-identification (re-ID) has drawn increasing attention in compute vision community because of its practical application prospects. Due to the inaccurate person detections and pose changes, pedestrian misalignment significantly increases the difficulty of feature extraction and matching. To address this problem, in this paper, we propose a Reference-Aided Part-Aligned (RAPA) framework to disentangle robust features of different parts. Firstly, in order to obtain better references between different videos, a pose-based reference feature learning module is introduced. Secondly, an effective relation-based part feature disentangling module is explored to align frames within each video. By means of using both modules, the informative parts of pedestrian in videos are well aligned and more discriminative feature representation is generated. Comprehensive experiments on three widely-used benchmarks, i.e. iLIDS-VID, PRID-2011 and MARS datasets verify the effectiveness of the proposed framework. Our code will be made publicly available.

As a prevailing task in video surveillance and forensics field, person re-identification (re-ID) aims to match person images captured from non-overlapped cameras. In unconstrained scenarios, person images often suffer from the resolution mismatch problem, i.e., Cross-Resolution Person Re-ID. To overcome this problem, most existing methods restore low resolution (LR) images to high resolution (HR) by super-resolution (SR). However, they only focus on the HR feature extraction and ignore the valid information from original LR images. In this work, we explore the influence of resolutions on feature extraction and develop a novel method for cross-resolution person re-ID called Multi-Resolution Representations Joint Learning (MRJL). Our method consists of a Resolution Reconstruction Network (RRN) and a Dual Feature Fusion Network (DFFN). The RRN uses an input image to construct a HR version and a LR version with an encoder and two decoders, while the DFFN adopts a dual-branch structure to generate person representations from multi-resolution images. Comprehensive experiments on five benchmarks verify the superiority of the proposed MRJL over the relevent state-of-the-art methods.