MV-ReID: 3D Multi-view Transformation Network for Occluded Person Re-Identification

Article Computer Science, Artificial Intelligence

Temporal Pixel-Level Semantic Understanding Through the VSPW Dataset

Jiaxu Miao et al.

Summary: This study introduces a large-scale video scene parsing dataset, VSPW, which contains 251,633 frames from 3,536 videos with pixel-level annotations. It covers 231 scenes and 124 object categories. The study also proposes a Temporal Attention Blending (TAB) network to improve pixel-level semantic understanding of videos. Experimental results demonstrate the superiority of the TAB approach on the VSPW dataset.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2023)

Add to Collection

Article Computer Science, Artificial Intelligence

ResT-ReID: Transformer block-based residual learning for person re-identification

Ying Chen et al.

Summary: This paper proposes a hybrid backbone network ResT-ReID based on ResNet-50 and Transformer block for person re-identification. It replaces depth-wise convolution in ResNet-50 with global self-attention and utilizes attention-guided Graph Convolution Networks to fully exploit the knowledge of individuals. Experimental results demonstrate that ResT-ReID achieves competitive results in person re-identification task.

PATTERN RECOGNITION LETTERS (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Cross-modality disentanglement and shared feedback learning for infrared-visible person re-identification

Kaifang Li et al.

Summary: This research proposes a novel cross-modality disentanglement and shared feedback learning framework for infrared-visible person re-identification. With two networks, the framework achieves modality-level and feature-level alignment while maintaining identity-consistency. Experimental results demonstrate that the method achieves competitive performance compared to state-of-the-art methods.

KNOWLEDGE-BASED SYSTEMS (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Semantic driven attention network with attribute learning for unsupervised person re-identification

Simin Xu et al.

Summary: Unsupervised domain adaptation for person re-identification addresses the challenges of cross-domain background variations and insufficient foreground identification. This paper presents a novel body structure estimation mechanism and an enforced semantic driven attention network, which improve feature representations through optimized attribute learning techniques.

KNOWLEDGE-BASED SYSTEMS (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Attention-calibration based double-branch cross-domain person re-identification

Kangkang Ji et al.

Summary: This paper proposes a new attention-calibration double-branch cross-domain pedestrian re-identification network, which achieves excellent performance in various environments by learning features from different domains and fusing them.

KNOWLEDGE-BASED SYSTEMS (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Dual Prototype Contrastive learning with Fourier Generalization for Domain Adaptive Person Re-identification

Xulin Song et al.

Summary: This paper proposes a Dual Prototype Contrastive learning with Fourier Generalization (DPCFG) framework for domain adaptive Person Re-identification. The DPCFG method bridges the domain gap using the Fourier Generalization strategy and fully exploits the hard positive instances within each cluster using the Dual Prototype Contrastive learning strategy. Experimental results demonstrate that DPCFG achieves state-of-the-art UDA person ReID performance.

KNOWLEDGE-BASED SYSTEMS (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Identifying Visible Parts via Pose Estimation for Occluded Person Re-Identification

Jiaxu Miao et al.

Summary: This article addresses the occlusion problem in person re-identification, proposing to incorporate pose information into the re-id framework to eliminate noise information.

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Learning Feature Recovery Transformer for Occluded Person Re-Identification

Boqiang Xu et al.

Summary: This paper proposes a new approach called FRT to address the challenges of occluded person re-identification. It utilizes visibility graph matching and feature recovery transformer to reduce noise interference during feature matching and recover complete features based on graph similarity. Extensive experiments demonstrate the effectiveness of FRT, outperforming state-of-the-art results on challenging datasets.

IEEE TRANSACTIONS ON IMAGE PROCESSING (2022)

Add to Collection

Article Computer Science, Information Systems

Hierarchical multi-view context modelling for 3D object classification and retrieval

An-An Liu et al.

Summary: Recent advances in 3D sensors and software have led to a large amount of 3D data. This paper proposes a hierarchical multi-view context modelling method to address the problem of learning discriminative multi-view visual characteristics in 3D object classification and retrieval. The proposed method demonstrates superiority in experiments on ModelNet10, ModelNet40 and ShapeNetCore55 datasets.

INFORMATION SCIENCES (2021)

Add to Collection

Review Computer Science, Hardware & Architecture

Review of multi-view 3D object recognition methods based on deep learning

Shaohua Qi et al.

Summary: This paper comprehensively reviews and classifies the latest developments in deep learning methods for multi-view 3D object recognition, summarizes results on mainstream datasets, provides insightful conclusions, and proposes enlightening future research directions.

DISPLAYS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence