3 papers
cs.CV2026
ORMOT: A Dataset and Framework for Omnidirectional Referring Multi-Object Tracking
Sijia Chen, Zihan Zhou, Yanqiu Yu +2
Multi-Object Tracking (MOT) is a fundamental task in computer vision, aiming to track targets across video frames. Existing MOT methods perform well in general visual scenes, but f…
cs.CV2025
Enhancing Self-Supervised Fine-Grained Video Object Tracking with Dynamic Memory Prediction
Zihan Zhou, Changrui Dai, Aibo Song +1
Successful video analysis relies on accurate recognition of pixels across frames, and frame reconstruction methods based on video correspondence learning are popular due to their e…
cs.CV2025
Leveraging Motion Information for Better Self-Supervised Video Correspondence Learning
Zihan Zhou, Changrui Dai, Aibo Song +1
Self-supervised video correspondence learning depends on the ability to accurately associate pixels between video frames that correspond to the same visual object. However, achievi…