A Simple Baseline for Pose Tracking in Videos of Crowded Scenes
It addresses pose tracking in complex, crowded environments for video analysis applications, but is incremental in approach.
The paper tackles pose tracking in crowded video scenes by combining multi-object tracking with pose estimation and optical flow, achieving competitive performance on a benchmark challenge.
This paper presents our solution to ACM MM challenge: Large-scale Human-centric Video Analysis in Complex Events\cite{lin2020human}; specifically, here we focus on Track3: Crowd Pose Tracking in Complex Events. Remarkable progress has been made in multi-pose training in recent years. However, how to track the human pose in crowded and complex environments has not been well addressed. We formulate the problem as several subproblems to be solved. First, we use a multi-object tracking method to assign human ID to each bounding box generated by the detection model. After that, a pose is generated to each bounding box with ID. At last, optical flow is used to take advantage of the temporal information in the videos and generate the final pose tracking result.