Die Dai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Die Dai has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Xu Zhang
Chongqing University of Posts and Telecommunications
2 shared publications - Hongwen Zhou
Chongqing Planning and Natural Resources Information Center
1 shared publication - Hongying Meng
Brunel University of London
1 shared publication - Jiaye Yang
University of Electronic Science and Technology of China, Chengdu, China
1 shared publication - Peiyuan Jiang
University of Electronic Science and Technology of China, Chengdu, China
1 shared publication - Qiao Liu
University of Electronic Science and Technology of China, Chengdu, China
1 shared publication
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 13175-13183
Region-Adaptive Modality Fusion with Semantic Expert Routing for Multi-Modal Crowd Counting
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 13175-13183
Open AccessRegion-Adaptive Modality Fusion with Semantic Expert Routing for Multi-Modal Crowd Counting
- Open Access
- 9 September 2025
- Published by Institution of Engineering and Technology (IET) in CAAI Transactions on Intelligence Technology
- Vol. 10 (5) , 1535-1547
A Keypoint‐Guided Feature Partition Network for Occluded Person Re‐Identification
Existing occluded person re‐identification methods employ hard or soft partition strategies to explore fine‐grained information. However, the hard partition strategy which extracts region‐level features may impair the semantic connectivity of correlated human body parts. A pose‐guided soft partition establishes correlations among human keypoints, while the generated pixel‐level embeddings may lose the surrounding semantic information. In this paper, we propose a keypoint‐guided feature partition (KGFP) method that consists of a feature extractor, a hard partition branch, and a soft partition branch. Specifically, we adopt a vision transformer and a pose estimator to extract features and keypoint information. In the hard partition branch, we partition features into distinct groups and classify them into nonoccluded, semi‐occluded, and occluded features to obtain region‐level features and filter out occlusions. Furthermore, we design a dissimilarity loss to reduce the similarity between semi‐occluded and occluded features. In the soft partition branch, we introduce a graph attention network and consider global and keypoint embeddings as nodes of a graph to discover interrelationships. Additionally, we formulate image alignment as a graph matching problem and propose a feature alignment‐based graph to reduce position misalignment. Extensive experiments demonstrate that the proposed method achieves superior performance compared to state‐of‐the‐art methods on Occluded‐DukeMTMC, Markt1501, and DukeMTMC‐reID.
- 9 September 2025
- Published by Institution of Engineering and Technology (IET) in CAAI Transactions on Intelligence Technology
- Vol. 10 (5) , 1535-1547
Open AccessA Keypoint‐Guided Feature Partition Network for Occluded Person Re‐Identification
Existing occluded person re‐identification methods employ hard or soft partition strategies to explore fine‐grained information. However, the hard partition strategy which extracts region‐level features may impair the semantic connectivity of correlated human body parts. A pose‐guided soft partition establishes correlations among human keypoints, while the generated pixel‐level embeddings may lose the surrounding semantic information. In this paper, we propose a keypoint‐guided feature partition (KGFP) method that consists of a feature extractor, a hard partition branch, and a soft partition branch. Specifically, we adopt a vision transformer and a pose estimator to extract features and keypoint information. In the hard partition branch, we partition features into distinct groups and classify them into nonoccluded, semi‐occluded, and occluded features to obtain region‐level features and filter out occlusions. Furthermore, we design a dissimilarity loss to reduce the similarity between semi‐occluded and occluded features. In the soft partition branch, we introduce a graph attention network and consider global and keypoint embeddings as nodes of a graph to discover interrelationships. Additionally, we formulate image alignment as a graph matching problem and propose a feature alignment‐based graph to reduce position misalignment. Extensive experiments demonstrate that the proposed method achieves superior performance compared to state‐of‐the‐art methods on Occluded‐DukeMTMC, Markt1501, and DukeMTMC‐reID.
- 3 June 2023
- Published by Springer Nature
- p. 61-67
Mixup Cluster Contrastive Learning Over Harder Samples for Unsupervised Person Re-identification
- 3 June 2023
- Published by Springer Nature
- p. 61-67
Mixup Cluster Contrastive Learning Over Harder Samples for Unsupervised Person Re-identification