Explicit Context Reasoning with Supervision for Visual Tracking
Authors:
- Fansheng ZengFZFansheng Zeng
- Bineng ZhongBZBineng Zhong
- Haiying XiaHXHaiying Xia
- Yufei TanYTYufei Tan
- Xiantao HuXHXiantao Hu
- Liangtao ShiLSLiangtao Shi
- Shuxiang SongSSShuxiang Song
Preprint
Abstract
Contextual reasoning with constraints is crucial for enhancing temporal consistency in cross-frame modeling for visual tracking. However, mainstream tracking algorithms typically associate context by merely stacking historical information without explicitly supervising the association process, making it difficult to effectively model the target's evolving dynamics. To alleviate this problem, we propose RSTrack, which explicitly models and supervises context reasoning via three core mechanisms. \textit{1) Context Reasoning Mechanism}: Constructs a target state reasoning pipeline, converting unconstrained contextual associations into a temporal reasoning process that predicts the current representation based on historical target states, thereby enhancing temporal consistency. \textit{2) Forward Supervision Strategy}: Utilizes true target features as anchors to constrain the reasoning pipeline, guiding the predicted output toward the true target distribution and suppressing drift in the context reasoning process. \textit{3) Efficient State Modeling}: Employs a compression-reconstruction mechanism to extract the core features of the target, removing redundant information across frames and preventing ineffective contextual associations. These three mechanisms collaborate to effectively alleviate the issue of contextual association divergence in traditional temporal modeling. Experimental results show that RSTrack achieves state-of-the-art performance on multiple benchmark datasets while maintaining real-time running speeds. Our code is available at https://github.com/GXNU-ZhongLab/RSTrack.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- Similarity-Guided Layer-Adaptive Vision Transformer for UAV Tracking2025
- Exploiting Multimodal Spatial-temporal Patterns for Video Object TrackingProceedings of the AAAI Conference on Artificial Intelligence, 2025
- Adaptive Perception for Unified Visual Multimodal Object TrackingIEEE Transactions on Artificial Intelligence, 2025
- Mamba Adapter: Efficient Multi-Modal Fusion for Vision-Language TrackingIEEE Transactions on Circuits and Systems for Video Technology, 2025
- Fast-iTPN: Integrally Pre-Trained Transformer Pyramid Network With Token MigrationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2024
- Toward Modalities Correlation for RGB-T TrackingIEEE Transactions on Circuits and Systems for Video Technology, 2024
- Transformer Tracking via Frequency FusionIEEE Transactions on Circuits and Systems for Video Technology, 2023
- Ocean: Object-Aware Anchor-Free Tracking2020
- SiamFC++: Towards Robust and Accurate Visual Tracking with Target Estimation GuidelinesProceedings of the AAAI Conference on Artificial Intelligence, 2020
- GOT-10k: A Large High-Diversity Benchmark for Generic Object Tracking in the WildIEEE Transactions on Pattern Analysis and Machine Intelligence, 2019