E^2TAD: An Energy-Efficient Tracking-based Action Detector
It addresses the need for efficient and precise action detection in applications like robotics and healthcare, though it appears incremental as it builds on existing tracking and detection paradigms.
The paper tackles the problem of fine-grained video action detection by proposing a tracking-based solution that accurately localizes actions spatially and temporally, achieving first place in the 2021 Low-Power Computer Vision Challenge.
Video action detection (spatio-temporal action localization) is usually the starting point for human-centric intelligent analysis of videos nowadays. It has high practical impacts for many applications across robotics, security, healthcare, etc. The two-stage paradigm of Faster R-CNN inspires a standard paradigm of video action detection in object detection, i.e., firstly generating person proposals and then classifying their actions. However, none of the existing solutions could provide fine-grained action detection to the "who-when-where-what" level. This paper presents a tracking-based solution to accurately and efficiently localize predefined key actions spatially (by predicting the associated target IDs and locations) and temporally (by predicting the time in exact frame indices). This solution won first place in the UAV-Video Track of 2021 Low-Power Computer Vision Challenge (LPCVC).