Refinements in Motion and Appearance for Online Multi-Object Tracking
arXiv:2003.07177
Abstract
Modern multi-object tracking (MOT) system usually involves separated modules, such as motion model for location and appearance model for data association. However, the compatible problems within both motion and appearance models are always ignored. In this paper, a general architecture named as MIF is presented by seamlessly blending the Motion integration, three-dimensional(3D) Integral image and adaptive appearance feature Fusion. Since the uncertain pedestrian and camera motions are usually handled separately, the integrated motion model is designed using our defined intension of camera motion. Specifically, a 3D integral image based spatial blocking method is presented to efficiently cut useless connections between trajectories and candidates with spatial constraints. Then the appearance model and visibility prediction are jointly built. Considering scale, pose and visibility, the appearance features are adaptively fused to overcome the feature misalignment problem. Our MIF based tracker (MIFT) achieves the state-of-the-art accuracy with 60.1 MOTA on both MOT16&17 challenges.
References in corpus (7)
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- Simple Online and Realtime Tracking
- MOT16: A Benchmark for Multi-Object Tracking
- MOTChallenge 2015: Towards a Benchmark for Multi-Target Tracking
- Multi-Object Tracking with Multiple Cues and Switcher-Aware Classification
- Temporal Dynamic Appearance Modeling for Online Multi-Person Tracking
- Tracklet Association Tracker: An End-to-End Learning-based Association Approach for Multi-Object Tracking