Online Multi-Object Tracking Using CNN-based Single Object Tracker with Spatial-Temporal Attention Mechanism
This work addresses computational efficiency and accuracy in multi-object tracking for video analysis, representing an incremental improvement with specific gains.
The paper tackles the problem of online multi-object tracking by proposing a CNN-based framework with a spatial-temporal attention mechanism to handle occlusion and drift, achieving 34.3% MOTA on MOT15 and 46.0% on MOT16 benchmarks.
In this paper, we propose a CNN-based framework for online MOT. This framework utilizes the merits of single object trackers in adapting appearance models and searching for target in the next frame. Simply applying single object tracker for MOT will encounter the problem in computational efficiency and drifted results caused by occlusion. Our framework achieves computational efficiency by sharing features and using ROI-Pooling to obtain individual features for each target. Some online learned target-specific CNN layers are used for adapting the appearance model for each target. In the framework, we introduce spatial-temporal attention mechanism (STAM) to handle the drift caused by occlusion and interaction among targets. The visibility map of the target is learned and used for inferring the spatial attention map. The spatial attention map is then applied to weight the features. Besides, the occlusion status can be estimated from the visibility map, which controls the online updating process via weighted loss on training samples with different occlusion statuses in different frames. It can be considered as temporal attention mechanism. The proposed algorithm achieves 34.3% and 46.0% in MOTA on challenging MOT15 and MOT16 benchmark dataset respectively.