Uncertainty-Aware Weakly Supervised Action Detection from Untrimmed Videos
This addresses the annotation cost problem for video action recognition researchers, though it is incremental as it builds on existing weakly-supervised and Multiple Instance Learning frameworks.
The paper tackles the problem of spatio-temporal action detection from untrimmed videos by using only video-level labels to avoid costly frame-by-frame annotation, achieving state-of-the-art results among weakly-supervised methods on UCF101-24 and reporting the first weakly-supervised results on the AVA dataset.
Despite the recent advances in video classification, progress in spatio-temporal action recognition has lagged behind. A major contributing factor has been the prohibitive cost of annotating videos frame-by-frame. In this paper, we present a spatio-temporal action recognition model that is trained with only video-level labels, which are significantly easier to annotate. Our method leverages per-frame person detectors which have been trained on large image datasets within a Multiple Instance Learning framework. We show how we can apply our method in cases where the standard Multiple Instance Learning assumption, that each bag contains at least one instance with the specified label, is invalid using a novel probabilistic variant of MIL where we estimate the uncertainty of each prediction. Furthermore, we report the first weakly-supervised results on the AVA dataset and state-of-the-art results among weakly-supervised methods on UCF101-24.