3C-Net: Category Count and Center Loss for Weakly-Supervised Action Localization
This addresses the problem of reducing annotation effort for action localization in videos, though it is incremental as it builds on existing weakly-supervised approaches.
The paper tackles weakly-supervised temporal action localization by proposing 3C-Net, which uses only video-level supervision with action category labels and counts, achieving a 4.6% absolute gain in mean average precision on THUMOS14 compared to state-of-the-art methods.
Temporal action localization is a challenging computer vision problem with numerous real-world applications. Most existing methods require laborious frame-level supervision to train action localization models. In this work, we propose a framework, called 3C-Net, which only requires video-level supervision (weak supervision) in the form of action category labels and the corresponding count. We introduce a novel formulation to learn discriminative action features with enhanced localization capabilities. Our joint formulation has three terms: a classification term to ensure the separability of learned action features, an adapted multi-label center loss term to enhance the action feature discriminability and a counting loss term to delineate adjacent action sequences, leading to improved localization. Comprehensive experiments are performed on two challenging benchmarks: THUMOS14 and ActivityNet 1.2. Our approach sets a new state-of-the-art for weakly-supervised temporal action localization on both datasets. On the THUMOS14 dataset, the proposed method achieves an absolute gain of 4.6% in terms of mean average precision (mAP), compared to the state-of-the-art. Source code is available at https://github.com/naraysa/3c-net.