Towards fast and kernelized orthogonal discriminant analysis on person re-identification

Two-phase self-supervised pretraining for object re-identification

Abstract. Person Re-Identification (Re-ID) is the task of matching the same person in multiple images captured by different cameras. Recently deep learning-based Re-ID algorithms demonstrated an exciting progress for terrestrial-based cameras, still, person Re-ID in aerial images poses multiple challenges including occlusion of human feature parts, image distortion, and dynamic camera location. In this paper, we propose a new Person Aerial Re-ID framework Robust to Occlusion and Thermal imagery (ParrotGAN). Our model is focused on cross-modality person Re-ID in aerial images. Furthermore, we collected a new large-scale synthetic multimodal AerialReID dataset with 30k images and 137 person ID. Our ParrotGAN model leverages two strategies to achieve robust performance in the task of person Re-ID in thermal and visible range. Firstly, we use a latent space of StyleGAN2 model to estimate the distance between two images of a person. Specifically, we project each real image into the latent space with a correspondent latent vector z. We use the distance between latent vectors to provide a Re-ID similarity metric. Secondly, we use a generative-adversarial network to translate a color image to a synthetic thermal image. We use synthetic image for a cross-modality Re-ID. We evaluate our ParrotGAN model and basleines on our AerialReID and PRAI-1581 datasets. The results of the evaluation are encouraging and demonstrate that our ParrotGAN model competes with baselines in visible range aerial person Re-ID and outperforms them in the cross-modality setting. We made our code and dataset publicly available.

Active learning has recently attracted increasing attention in the task of person re-identification, due to its unique scalability that not only maximally reduces the annotation cost but also retains the satisfying performance. Although some preliminary active learning methods have been explored in scalable person re-identification task, they have the following two problems: 1) the inefficiency in the selection process of image pairs due to the huge search space, and 2) the ineffectiveness caused by ignoring the impact of unlabeled data in model training. Considering that, we propose a Multi-grained Active Semi-Supervised learning framework, named MASS, to address the scalable person re-identification problem existing in the practical scenarios. Specifically, we firstly design a cluster-scatter procedure to alleviate the inefficiency problem, which consists of two components: cluster step and scatter step. The cluster step shrinks the search space into individual small clusters by a coarse-grained clustering method, and the subsequent scatter step further mines the hard distinguished image pairs from unlabelled set to purify the learned clusters by a novel centrality-based adaptive purification strategy. Afterward, we introduce a customized purification loss for the purified clustering, which utilizes the complementary information in both labeled and unlabeled data to optimize the model for solving the ineffectiveness problem. The cluster-scatter procedure and the model optimization are performed in an iterative fashion to achieve the promising performance while greatly reducing the annotation cost. Extensive experimental results have demonstrated that MASS can even achieve a competitive performance with fully supervised methods in the case of extremely less annotation requirements.

Compact feature representation of person image is important for person re-identification (Re-ID) task. Recently, part-based representation models have been widely studied for extracting the more compact and robust feature representation for person image to improve person Re-ID results. However, existing part-based representation models mostly extract the features of different parts independently which ignore the spatial relationship information among different parts. To address this issue, in this paper we propose a novel deep learning framework, named Part-based Hierarchical Graph Convolutional Network (PH-GCN) for person Re-ID problem. Given a person image, PH-GCN first constructs a hierarchical graph to represent the spatial relationships among different parts. Then, both local and global feature learning is achieved by the feature information passing in PH-GCN, which takes the information of other parts into account for part feature representation. Finally, a perceptron layer is adopted for the final person part label prediction and re-identification. The proposed framework provides a general solution that integrates <i>local</i>, <i>global</i> and <i>structural</i> feature learning simultaneously in a unified end-to-end network representation and learning. Extensive experiments on several widely used benchmark datasets demonstrate the effectiveness and benefits of the proposed PH-GCN approach for person Re-ID task.

Deep learning has shown significant successes in person reidentification (re-id) tasks. However, most existing works focus on discriminative feature learning and impose complex neural networks, suffering from low inference efficiency. In fact, feature extraction time is also crucial for real-world applications and lightweight models are needed. Prevailing pruning methods usually pay attention to compact classification models. However, these methods are suboptimal for compacting re-id models, which usually produce continuous features and are sensitive to network pruning. The key point of pruning re-id models is how to retain the original filter distribution in continuous features as much as possible. In this work, we propose a blockwise adjacent filter decaying method to fill this gap. Specifically, given a trained model, we first evaluate the redundancy of filters based on the adjacency relationships to preserve the original filter distribution. Second, previous layerwise pruning methods ignore that discriminative information is enhanced block-by-block. Therefore, we propose a blockwise filter pruning strategy to better utilize the block relations in the pretrained model. Third, we propose a novel filter decaying policy to progressively reduce the scale of redundant filters. Different from conventional soft filter pruning that directly sets the filter values as zeros, the proposed filter decaying can keep the pretrained knowledge as much as possible. We evaluate our method on three popular person reidentification datasets, that is: 1) Market-1501; 2) DukeMTMC-reID; and 3) MSMT17_V1. The proposed method shows superior performance to the existing state-of-the-art pruning methods. After pruning over 91.9% parameters on DukeMTMC-reID, the Rank-1 accuracy only drops 3.7%, demonstrating its effectiveness for compacting person reidentification.

Recently, the person re-identification task becomes increasingly crucial in crowded scenarios, e.g., airports and schools. Many methods with high performance have been proposed to solve this problem. However, the existence of occlusion still challenges the development of person re-identification. In this work, we present the novel Pose - Guided Mixed Attention Network (PGMANet), an end-to-end framework to deal with pedestrian reidentification under occluded situations by fusing posture and second-order information. Specially, we employ two models. The first is Human Part - level Attention Model. We use key point information of a pedestrian to generate a heat map to enhance the pedestrian body part's feature. Simultaneously, we design Second - order Information Attention Model to investigate the correlation among features of different parts. Experimental results show that our method achieves state-of-the-art person re-identification performance on two challenging occlusion datasets Occluded-DukeMTMC and Occluded-Reid.

Matching people across non-overlapping camera views, known as person re-identification, is challenging due to the lack of spatial and temporal constraints and large visual appearance changes caused by variations in view angle, lighting, background clutter and occlusion. To address these challenges, most previous approaches aim to extract visual features that are both distinctive and stable under appearance changes. However, most visual features and their combinations under realistic conditions are neither stable nor distinctive thus should not be used indiscriminately. In this paper, we propose to formulate person re-identification as a distance learning problem, which aims to learn the optimal distance that can maximises matching accuracy regardless the choice of representation. To that end, we introduce a novel Probabilistic Relative Distance Comparison (PRDC) model, which differs from most existing distance learning methods in that, rather than minimising intra-class variation whilst maximising intra-class variation, it aims to maximise the probability of a pair of true match having a smaller distance than that of a wrong match pair. This makes our model more tolerant to appearance changes and less susceptible to model over-fitting. Extensive experiments are carried out to demonstrate that 1) by formulating the person re-identification problem as a distance learning problem, notable improvement on matching accuracy can be obtained against conventional person re-identification techniques, which is particularly significant when the training sample size is small; and 2) our PRDC outperforms not only existing distance learning methods but also alternative learning methods based on boosting and learning to rank.

This paper proposes a novel image representation which can properly handle both background and illumination variations. It is therefore adapted to the person/face reidentification tasks, avoiding the use of any additional pre-processing steps such as foreground-background separation or face and body part segmentation. This novel representation relies on the combination of Biologically Inspired Features (BIF) and covariance descriptors used to compute the similarity of the BIF features at neighboring scales. Hence, we will refer to it as the BiCov representation. To show the effectiveness of BiCov, this paper conducts experiments on two person re-identification tasks (VIPeR and ETHZ) and one face verification task (LFW), on which it improves the current state-of-the-art performance.

Prevalent nighttime person re-identification (ReID) methods typically combine image relighting and ReID networks in a sequential manner. However, their performance (recognition accuracy) is limited by the quality of relighting images and insufficient collaboration between image relighting and ReID tasks. To handle these problems, we propose a novel Collaborative Enhancement Network called CENet, which performs the multilevel feature interactions in a parallel framework, for nighttime person ReID. In particular, the designed parallel structure of CENet can not only avoid the impact of the quality of relighting images on ReID performance, but also allow us to mine the collaborative relations between image relighting and person ReID tasks. To this end, we integrate the multilevel feature interactions in CENet, where we first share the Transformer encoder to build the low-level feature interaction, and then perform the feature distillation that transfers the high-level features from image relighting to ReID, thereby alleviating the severe image degradation issue caused by the nighttime scenario while avoiding the impact of relighting images. In addition, the sizes of existing real-world nighttime person ReID datasets are limited, and large-scale synthetic ones exhibit substantial domain gaps with real-world data. To leverage both small-scale real-world and large-scale synthetic training data, we develop a multi-domain learning algorithm, which alternately utilizes both kinds of data to reduce the inter-domain difference in training procedure. Extensive experiments on two real nighttime datasets, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Night600</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">RGBNT201_rgb</i> , and a synthetic nighttime ReID dataset are conducted to validate the effectiveness of CENet. We release the code and synthetic dataset at: https://github.com/Alexadlu/CENet.

Unsupervised domain adaptation (UDA) in the task of person re-identification (re-ID) is highly challenging due to large domain divergence and no class overlap between domains. Pseudo-label based self-training is one of the representative techniques to address UDA. However, label noise caused by unsupervised clustering is always a trouble to self-training methods. To depress noises in pseudo-labels, this paper proposes a Noise Resistible Mutual-Training (NRMT) method, which maintains two networks during training to perform collaborative clustering and mutual instance selection. On one hand, collaborative clustering eases the fitting to noisy instances by allowing the two networks to use pseudo-labels provided by each other as an additional supervision. On the other hand, mutual instance selection further selects reliable and informative instances for training according to the peer-confidence and relationship disagreement of the networks. Extensive experiments demonstrate that the proposed method outperforms the state-of-the-art UDA methods for person re-ID.

As a fundamental task in automated video surveillance, person re-identification, which has received increasing attention in recent years, aims to match people across non-overlapping camera views in a multi-camera surveillance system. It has been reported that KISS metric learning has been followed by most of the previous supervised work because of its state of the art performance for person re-identification on VIPeR dataset. However, given only a small number of labeled image pairs available for training, the matching model certainly suffers from unstable learning process and poor matching result. To address this serious practical issue, we proposed a novel semi-supervised KISS metric learning (SS-KISS) approach which makes use of unlabeled data to improve the re-identification performance by 1) combining both global and local information to select the most confident image pairs from the unlabeled data; 2) using an ensemble approach, which explores advantages of supervised and unsupervised learning by reconciling two matching models on which labeled and un-labeled data to an optimal one via smart weighting schema. Extensive experiments have been conducted on three datasets: VIPeR, ETHZ, and i-LiDS, experimental results demonstrate that our approach achieves a sound performance in the case of small amount of labeled data.

With the rise of command city construction, the importance of pedestrian re-identification based on video surveillance has become increasingly prominent. The person re-identification task mainly studies recognizing images of the same person under different cameras. Due to the high cost of labeling data in supervised learning in common methods, and the lack of guidance for labeling data in unsupervised learning, the experimental results are unsatisfactory. In this paper, we propose a new weakly supervised person re-identification problem framework: a human-computer interaction-based sample active push and auxiliary labeling method is designed to quickly obtain a small number of identity-related labels with low labor cost; At the same time, the weak supervision information that is easily obtained in the real scene is introduced into the pseudo-label generation process to improve the clustering method, aiming to solve the problem of insufficient initial labeling data of images in the label estimation stage and weakly supervised labels are not well utilized; Finally, in view of the insufficient utilization of weakly supervised labels in the model training process, a weakly supervised loss function WSP loss is designed and added to the training to optimize the overall loss function. Compared with the current state-of-the-art unsupervised and weakly supervised person re-identification work, this paper shows better performance under the same conditions and even using less labeled information.

Extracting robust feature representation is one of the key challenges for person re‐identification (ReID) task. Although convolution neural network (CNN)‐based methods have achieved great success, they still cannot handle the part occlusion and misalignment caused by limited receptive field. Recently, pure transformer models have shown its power in the person ReID task. However, current transformer models adopt patches of equal‐scale as input, and cannot solve the problem of cross‐scale interaction properly. To overcome this problem, an adaptive cross‐scale transformer from a perspective of the graph signal, named ACSFormer, is proposed. Specifically, the self‐attention module is first treated as an undirected fully connected graph. And then, “node variation” is introduced as an indicator to adaptively merge neighbourhood tokens. To the best of the authors’ knowledge, their ACSFormer is the first work to attempt to combine pure transformers and graph signal processing in the field of person ReID. Extensive evaluations are conducted on three person ReID datasets to validate the performance of ACSFormer. Experiments demonstrate that this ACSFormer performs on par with state‐of‐the‐art CNN‐based methods and consistently improves transformer‐based baseline, for example, surpassing ViT‐baseline by 2.5%, 2.7% and 4.8% mAP on Market1501, DukeMTMC‐reID and MSMT17, respectively.

A review on video person re-identification based on deep learning

Person re-identification is an important task of matching pedestrians across non-overlapping camera views. In this paper, we exploit a weighted feature descriptor for person re-identification. We firstly compute the weights on the superpixel level via graph-based manifold ranking algorithm, then integrate the computed weights into a patch-based feature descriptor, named local maximal occurrence. Finally, the weighted descriptors are fed into a top-push distance learning to mitigate the cross-view gaps. We evaluate the proposed method on three benchmark datasets iLIDS-VID, PRID 450S and VIPeR. The promising experimental results demonstrate the effectiveness of the proposed method comparing with the state-of-the-arts.