Feature Erasing and Diffusion Network for Occluded Person Re-Identification

Wang, Zhikang; Zhu, Feng; Tang, Shixiang; Zhao, Rui; He, Lihuo; Song, Jiangning

doi:10.1109/cvpr52688.2022.00471

Cited by 115 publications

(47 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the P-DukeMTMC dataset, our FCFormer outperforms QPM by 1.0% and 8.3% in terms of Rank-1 accuracy and mAP respectively. On the Occluded-REID dataset, FCFormer produces the comparable results with FED [28], achieving 84.9% Rank-1. We fail to achieve the highest Rank-1 performance on Occluded-REID for the following reasons: First, the Transformer has poor cross-domain generalization ability on small datasets.…”

Section: Performance Under Transfer Settingmentioning

confidence: 89%

“…Table II shows the results on three occluded datasets, i.e., Occluded-Duke and P-DukeMTMC. As table shows, two classes of methods will be compared: CNN based ReID methods [3], [4], [6], [7], [10], [13], [16], [29]- [31], [33]- [39], [42]- [44] and Transformer based ReID methods [5], [20], [21], [23], [28]. The results show that the proposed FCFormer consistently achieves the competitive performance on Occluded datasets.…”

Section: Comparison With the State-of-the-art Modelsmentioning

confidence: 99%

See 1 more Smart Citation

Pose-Guided Feature Disentangling for Occluded Person Re-identification Based on Transformer

Wang

Liu

Song

et al. 2022

AAAI

115

View full text Add to dashboard Cite

Occluded person re-identification is a challenging task as human body parts could be occluded by some obstacles (e.g. trees, cars, and pedestrians) in certain scenes. Some existing pose-guided methods solve this problem by aligning body parts according to graph matching, but these graph-based methods are not intuitive and complicated. Therefore, we propose a transformer-based Pose-guided Feature Disentangling (PFD) method by utilizing pose information to clearly disentangle semantic components (e.g. human body or joint parts) and selectively match non-occluded parts correspondingly. First, Vision Transformer (ViT) is used to extract the patch features with its strong capability. Second, to preliminarily disentangle the pose information from patch information, the matching and distributing mechanism is leveraged in Pose-guided Feature Aggregation (PFA) module. Third, a set of learnable semantic views are introduced in transformer decoder to implicitly enhance the disentangled body part features. However, those semantic views are not guaranteed to be related to the body without additional supervision. Therefore, Pose-View Matching (PVM) module is proposed to explicitly match visible body parts and automatically separate occlusion features. Fourth, to better prevent the interference of occlusions, we design a Pose-guided Push Loss to emphasize the features of visible body parts. Extensive experiments over five challenging datasets for two tasks (occluded and holistic Re-ID) demonstrate that our proposed PFD is superior promising, which performs favorably against state-of-the-art methods. Code is available at https://github.com/WangTaoAs/PFD_Net

show abstract

Section: Performance Under Transfer Settingmentioning

confidence: 89%

Section: Comparison With the State-of-the-art Modelsmentioning

confidence: 99%

Pose-Guided Feature Disentangling for Occluded Person Re-identification Based on Transformer

Wang

Liu

Song

et al. 2022

AAAI

115

View full text Add to dashboard Cite

show abstract

“…However, the incomplete information caused by occlusion increases its difficulty significantly. Recent works [25] [36] attempted to solve the occlusion problem by utilizing pose information provided by the estimation algorithm as crucial supplementary guidance. Miao et al [25] proposed a pose-guided feature alignment (PGFA) method that employs pose key-points information to generate the attention map to direct discriminative feature learning.…”

Section: Occluded Person Re-identificationmentioning

confidence: 99%

Occluded Person Re-Identification With Pose Estimation Correction and Feature Reconstruction

Liu

Wang

et al. 2023

IEEE Access

View full text Add to dashboard Cite

In the real-world surveillance system application, the accuracy of person re-identification (Re-ID) still suffers from occlusion. Occluded Re-ID aims to retrieve occluded person images from multiple cameras that do not overlap. To address this issue, we present a multi-branch feature enhancement network in this paper. It has the pose correction module (PCM), the feature reconstruction module (FRM), and the part align module (PAM). PCM builds the key-points confidence corrected mechanism of the self-adaptive weights learning before the part matching stage to strengthen part-level features in the non-occluded region and weaken ones in the occluded region. FRM reconstructs the feature distribution of inter-domain and class-domain using the maximum mean discrepancy to generalize the global features. The proposed feature separation mechanism enables the two branches to attend to distinct pedestrian feature. To increase the power of representation, PAM combines part-level and global features in an outer product. The experimental results on three holistic datasets, two partial datasets, and two occluded datasets demonstrate that our method is superior. It significantly outperforms state-of-the-art by 5.7% and 4.4% rank-1 scores on Occluded-Duke and Occluded-REID, respectively.

show abstract

“…Consequently, these methods significantly degrade when dealing with occluded images. While recent endeavors have facilitated person Re-ID under occlusion conditions [39,36,31,18,28,16], two main problems associated with occlusions still need to be addressed. Firstly, the presence of obstacles will vanish some parts of the human body, missing and misaligned extracted features.…”

Section: Introductionmentioning

confidence: 99%

“…Recently, with the exploration of attention mechanisms for various vision tasks, it has also been adopted for occluded person Re-ID to eliminate the interference of noisy information [44,29,12]. During the process of attention learning, many data augmentation strategies [1,36,51] generate artificial occlusion, which directs the attention to person and forces it to avoid occluded regions. Currently, the most widely used artificial occlusion methods are random erasing [48] or using the background as occlusion [1].…”

Section: Introductionmentioning

confidence: 99%

Attention Disturbance and Dual-Path Constraint Network for Occluded Person Re-Identification

Xia¹,

Tan²,

Dai³

et al. 2023

Preprint

View full text Add to dashboard Cite

Occluded person re-identification (Re-ID) aims to address the potential occlusion problem when matching occluded or holistic pedestrians from different camera views. Many methods use the background as artificial occlusion and rely on attention networks to exclude noisy interference. However, the significant discrepancy between simple background occlusion and realistic occlusion can negatively impact the generalization of the network. To address this issue, we propose a novel transformer-based Attention Disturbance and Dual-Path Constraint Network (ADP) to enhance the generalization of attention networks. Firstly, to imitate real-world obstacles, we introduce an Attention Disturbance Mask (ADM) module that generates an offensive noise, which can distract attention like a realistic occluder, as a more complex form of occlusion. Secondly, to fully exploit these complex occluded images, we develop a Dual-Path Constraint Module (DPC) that can obtain preferable supervision information from holistic images through dualpath interaction. With our proposed method, the network can effectively circumvent a wide variety of occlusions using the basic ViT baseline. Comprehensive experimental evaluations conducted on person re-ID benchmarks demonstrate the superiority of ADP over state-of-the-art methods.

show abstract

Feature Erasing and Diffusion Network for Occluded Person Re-Identification

Cited by 115 publications

References 71 publications

Pose-Guided Feature Disentangling for Occluded Person Re-identification Based on Transformer

Pose-Guided Feature Disentangling for Occluded Person Re-identification Based on Transformer

Occluded Person Re-Identification With Pose Estimation Correction and Feature Reconstruction

Attention Disturbance and Dual-Path Constraint Network for Occluded Person Re-Identification

Contact Info

Product

Resources

About