Not only Look, but also Listen: Learning Multimodal Violence Detection under Weak Supervision
This work addresses violence detection for security and content moderation by providing a new dataset and method, but it is incremental as it builds on existing multimodal and weak supervision approaches.
The authors tackled violence detection in videos by releasing a large-scale multimodal dataset (XD-Violence) and proposing a neural network with three parallel branches to capture relationships among video snippets, achieving state-of-the-art performance on their dataset and other benchmarks.
Violence detection has been studied in computer vision for years. However, previous work are either superficial, e.g., classification of short-clips, and the single scenario, or undersupplied, e.g., the single modality, and hand-crafted features based multimodality. To address this problem, in this work we first release a large-scale and multi-scene dataset named XD-Violence with a total duration of 217 hours, containing 4754 untrimmed videos with audio signals and weak labels. Then we propose a neural network containing three parallel branches to capture different relations among video snippets and integrate features, where holistic branch captures long-range dependencies using similarity prior, localized branch captures local positional relation using proximity prior, and score branch dynamically captures the closeness of predicted score. Besides, our method also includes an approximator to meet the needs of online detection. Our method outperforms other state-of-the-art methods on our released dataset and other existing benchmark. Moreover, extensive experimental results also show the positive effect of multimodal (audio-visual) input and modeling relationships. The code and dataset will be released in https://roc-ng.github.io/XD-Violence/.