PERSONAL ACADEMIC HOMEPAGE
About Me
I am a Ph.D. student at Anhui University . My research centers on unaligned RGB-T tracking and pedestrian attribute recognition (PAR). I study how to connect visual information across sensors and how to recognize people from rich semantic attributes.
In RGB-T tracking, I work on cross-modal alignment, fusion, and robust tracking when visible and thermal images do not line up. In PAR, I explore vision-language learning, structured prediction, and cross-domain benchmarks.
01 / RESEARCH
Research Interests
Unaligned RGB-T Tracking Cross-modal correspondence, progressive alignment, and reliable visible–thermal fusion for object tracking.
Pedestrian Attribute Recognition Vision-language modeling, attribute generation, and evaluation across diverse domains.
02 / OPEN SOURCE
Open-source Collections
2026 Our AALTrack and cross-modal UAV tracking and benchmark papers appeared in IEEE TCSVT .
2026 PMATrack appeared at CVPR 2026 . I also co-authored CVPR 2026 papers on modality-missing RGBT tracking and RGB-Event pedestrian attribute recognition.
2026 Our unaligned UAV RGB-T tracking benchmark and approach appeared at AAAI 2026 .
2026 SequencePAR appeared in Pattern Recognition , volume 172. MSP60K / LLM-PAR appeared at AAAI 2025 .
Selected peer-reviewed papers in RGBT tracking and pedestrian attribute recognition. First-author roles are marked, and my name is highlighted in each author list.
All 9
RGB-T Tracking 5
PAR 4
RGB + THERMAL
CVPR 2026 FIRST AUTHOR
Progressive Multi-cue Alignment for Unaligned RGBT Tracking
Jiandong Jin , Chenglong Li, Hao Feng, Andong Lu, Lili Huang, Jin Tang
Progressively aligns visible and thermal views to support robust tracking without pre-aligned inputs.
ATTRIBUTE GENERATION
Pattern Recognition 2026 FIRST AUTHOR
SequencePAR: Understanding Pedestrian Attributes via a Sequence Generation Paradigm
Jiandong Jin , Xiao Wang, Yin Lin, Chenglong Li, Lili Huang, Aihua Zheng, Jin Tang
Recasts pedestrian attributes as a sequence generation problem with visual and language cues.
MSP60K + LLM-PAR
AAAI 2025 FIRST AUTHOR
Pedestrian Attribute Recognition: A New Benchmark Dataset and a Large Language Model Augmented Framework
Jiandong Jin , Xiao Wang, Qian Zhu, Haiyang Wang, Chenglong Li
Presents the cross-domain MSP60K dataset and an LLM-augmented recognition framework.
RGB + THERMAL
IEEE TCSVT 2026 CO-AUTHOR
Efficient RGBT Tracking via Alternating Attention Learning
Lili Huang, Chengzhi Yan, Jiandong Jin , Lei Liu, Chenglong Li
Alternates attention prediction and full self-attention across RGB and thermal branches for efficient tracking.
CM-UOT BENCHMARK
IEEE TCSVT 2026 CO-AUTHOR
Cross-Modal UAV Object Tracking: State-Aware Representation Learning and A Unified Benchmark
Yun Xiao, Zhihong Hong, Jiandong Jin , Chenglong Li, Jin Tang, Amir Hussain
Introduces the CM-UOT cross-modal UAV benchmark and state-aware representation learning for modality switches.
RELATED RGBT TRACKING
CVPR 2026 CO-AUTHOR
Spatio-Temporal Conditional Denoising Transformer for Modality-Missing RGBT Tracking
Andong Lu, Ziyi Zha, Jiandong Jin , Shihao Li, Chenglong Li, Jin Tang, Bin Luo
Reconstructs missing or weak modality features with spatial and temporal cues for robust RGBT tracking.
UAV BENCHMARK
AAAI 2026 CO-AUTHOR
Unaligned UAV RGBT Tracking: A Largescale Benchmark and a Novel Approach
Yun Xiao, Yuhang Wang, Jiandong Jin , Wankang Zhang, Chenglong Li
Introduces the LUART benchmark and a tracking approach for unaligned UAV visible–thermal video.
RGB + EVENT PAR
CVPR 2026 CO-AUTHOR
RGB-Event based Pedestrian Attribute Recognition: A Benchmark Dataset and An Asymmetric RWKV Fusion Framework
Xiao Wang, Haiyang Wang, Shiao Wang, Qiang Chen, Jiandong Jin , Haoyu Song, Bo Jiang, Chenglong Li
Introduces the EventPAR benchmark and an asymmetric RWKV framework for RGB-Event pedestrian attribute recognition.
VISION + LANGUAGE
IEEE TCSVT 2025 CO-AUTHOR
Pedestrian Attribute Recognition via CLIP based Prompt Vision-Language Fusion
Xiao Wang, Jiandong Jin , Chenglong Li, Jin Tang, Cheng Zhang, Wei Wang
Uses CLIP and prompt tuning to connect pedestrian regions with attribute descriptions.
Readers around the world
Live visitor counts and approximate locations by country. Counting began on September 30, 2026.
View full statistics ↗
© 2026 Jiandong Jin Built for clarity and easy updates · Source ↗