Multi-Stream Single Shot Spatial-Temporal Action Detection
This work addresses action detection in video for computer vision applications, representing an incremental improvement by combining existing techniques like 3D CNN and SSD in a novel way.
The paper tackles spatial-temporal action detection in videos by proposing a 3D CNN-based single-shot detector that integrates short-term appearance and motion streams with long-term context streams, achieving a state-of-the-art frame-mAP of 71.30% on the UCF101-24 dataset for one-stage methods.
We present a 3D Convolutional Neural Networks (CNNs) based single shot detector for spatial-temporal action detection tasks. Our model includes: (1) two short-term appearance and motion streams, with single RGB and optical flow image input separately, in order to capture the spatial and temporal information for the current frame; (2) two long-term 3D ConvNet based stream, working on sequences of continuous RGB and optical flow images to capture the context from past frames. Our model achieves strong performance for action detection in video and can be easily integrated into any current two-stream action detection methods. We report a frame-mAP of 71.30% on the challenging UCF101-24 actions dataset, achieving the state-of-the-art result of the one-stage methods. To the best of our knowledge, our work is the first system that combined 3D CNN and SSD in action detection tasks.