Cross-modal Consensus Network for Weakly Supervised Temporal Action Localization
arXiv:2107.12589
Abstract
Weakly supervised temporal action localization (WS-TAL) is a challenging task that aims to localize action instances in the given video with video-level categorical supervision. Both appearance and motion features are used in previous works, while they do not utilize them in a proper way but apply simple concatenation or score-level fusion. In this work, we argue that the features extracted from the pretrained extractor, e.g., I3D, are not the WS-TALtask-specific features, thus the feature re-calibration is needed for reducing the task-irrelevant information redundancy. Therefore, we propose a cross-modal consensus network (CO2-Net) to tackle this problem. In CO2-Net, we mainly introduce two identical proposed cross-modal consensus modules (CCM) that design a cross-modal attention mechanism to filter out the task-irrelevant information redundancy using the global information from the main modality and the cross-modal local information of the auxiliary modality. Moreover, we treat the attention weights derived from each CCMas the pseudo targets of the attention weights derived from another CCM to maintain the consistency between the predictions derived from two CCMs, forming a mutual learning manner. Finally, we conduct extensive experiments on two common used temporal action localization datasets, THUMOS14 and ActivityNet1.2, to verify our method and achieve the state-of-the-art results. The experimental results show that our proposed cross-modal consensus module can produce more representative features for temporal action localization.
arXiv commentACM International Conference on Multimedia, 2021
References in corpus (17)
- Adam: A Method for Stochastic Optimization
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- The Kinetics Human Action Video Dataset
- Temporal Action Localization in Untrimmed Videos via Multi-stage CNNs
- Temporal Action Detection with Structured Segment Networks
- BSN: Boundary Sensitive Network for Temporal Action Proposal Generation
- Activity Graph Transformer for Temporal Action Localization
- Weakly-Supervised Action Localization with Expectation-Maximization Multi-Instance Learning
- 3C-Net: Category Count and Center Loss for Weakly-Supervised Action Localization
- Learning Cross-Modal Deep Representations for Robust Pedestrian Detection
- Weakly-Supervised Action Localization by Generative Attention Modeling
- MIST: Multiple Instance Self-Training Framework for Video Anomaly Detection
- Two-Stream Consensus Network for Weakly-Supervised Temporal Action Localization
- SF-Net: Single-Frame Supervision for Temporal Action Localization
- ACSNet: Action-Context Separation Network for Weakly Supervised Temporal Action Localization
- A Local-to-Global Approach to Multi-modal Movie Scene Segmentation
- MINI-Net: Multiple Instance Ranking Network for Video Highlight Detection