Real-time Online Action Detection Forests using Spatio-temporal Contexts
This addresses the problem of real-time action detection for applications requiring immediate processing, though it is incremental as it builds on existing skeleton-based random forest classifiers.
The paper tackles real-time online action detection by proposing a random forest framework that uses spatio-temporal contexts from CNN features during training to improve accuracy, achieving significant accuracy gains over state-of-the-art methods while maintaining real-time efficiency on datasets like MSRAction3D, G3D, and OAD.
Online action detection (OAD) is challenging since 1) robust yet computationally expensive features cannot be straightforwardly used due to the real-time processing requirements and 2) the localization and classification of actions have to be performed even before they are fully observed. We propose a new random forest (RF)-based online action detection framework that addresses these challenges. Our algorithm uses computationally efficient skeletal joint features. High accuracy is achieved by using robust convolutional neural network (CNN)-based features which are extracted from the raw RGBD images, plus the temporal relationships between the current frame of interest, and the past and future frames. While these high-quality features are not available in real-time testing scenario, we demonstrate that they can be effectively exploited in training RF classifiers: We use these spatio-temporal contexts to craft RF's new split functions improving RFs' leaf node statistics. Experiments with challenging MSRAction3D, G3D, and OAD datasets demonstrate that our algorithm significantly improves the accuracy over the state-of-the-art online action detection algorithms while achieving the real-time efficiency of existing skeleton-based RF classifiers.