Dong Wang avatar

汪栋

Dong Wang

Ph.D. Student

National University of Defense Technology

3D Understanding Large-scale Pre-training Multimodal Perception Re-identification
dongwang25@nudt.edu.cn Google Scholar

Publications

5 papers

SpecBridge: Spectral Graph Bridging for 3D-2D-Text Pre-training

Dong Wang , Jie Jiang , Weidong Min , Lixin Zhan , Xinpeng Zhao , Ze Zhang

International Joint Conference on Artificial Intelligence 2026 (IJCAI2026)

A 3D-2D-Text pre-training framework for open-vocabulary 3D understanding.

3D Understanding Vision-Language Learning Multi-Modal Learning CCF-B International Conference

VehicleMAE: Masked Autoencoding for Robust Vehicle Re-Identification

Qi Wang , Zeyu Zhang , Dong Wang , Di Gai , Xin Xiong , Jiyang Xu , Ruihua Zhou

International Conference on Computer Vision 2025 (ICCV 2025)

A masked autoencoding approach for learning robust vehicle representations under challenging visual conditions.

Vehicle Re-ID Autonomous Driving Large-scale Pre-training CCF-A International Conference

Threefold Encoder Interaction: Hierarchical Multi-Grained Semantic Alignment for Cross-Modal Food Retrieval

Qi Wang , Dong Wang , Weidong Min , Di Gai , Qing Han , Cheng Zha , Yuling Zhong

IEEE Transactions on Multimedia (TMM)

A threefold encoder interaction framework for cross-modal food retrieval with hierarchical multi-grained semantic alignment.

Cross-Modal Retrieval Food Computing Vision-Language Learning Multimodal Learning CCF-A Journal

Vision-Language Constraint Graph Representation Learning for Unsupervised Vehicle Re-identification

Dong Wang , Qi Wang , Zhiwei Tu , Weidong Min , Xin Xiong , Yuling Zhong , Di Gai

Expert Systems With Applications (ESWA)

A vision-language constraint graph representation learning framework for unsupervised vehicle Re-ID.

Vehicle Re-ID Vision-Language Learning Graph Representation Learning Unsupervised Learning SCI-Q1 Journal

SAM-driven MAE Pre-training and Background-aware Meta-learning for Unsupervised Vehicle Re-identification

Dong Wang , Qi Wang , Weidong Min , Di Gai , Qing Han , Longfei Li , Yuhan Geng

Computational Visual Media 2024 (CVM2024)

A SAM-driven MAE pre-training and background-aware meta-learning framework for reducing background interference in unsupervised vehicle Re-ID.

Vehicle Re-ID Unsupervised Learning Self-Supervised Learning Meta-Learning CCF-B Conference