2017
AMTnet: Action-Micro-Tube Regression by End-To-End Trainable Deep Architecture
ICCV 2017poster
Dominant approaches to action detection can only provide sub-optimal solutions to the problem, as they rely on seeking frame-level detections, to later compose them into "action tubes" in a post-processing step. With this paper we radically depart from current practice, and take a first step towards…