SAR-NAS: Skeleton-based Action Recognition via Neural Architecture Searching
This work addresses the need for more compact and efficient networks in skeleton-based action recognition, though it is incremental as it builds on existing DARTS methodology.
The paper tackled the problem of automatically designing neural network architectures for skeleton-based action recognition, resulting in a method that searches for compact networks achieving comparable or better performance than state-of-the-art methods on NTU RGB+D and Kinectics datasets.
This paper presents a study of automatic design of neural network architectures for skeleton-based action recognition. Specifically, we encode a skeleton-based action instance into a tensor and carefully define a set of operations to build two types of network cells: normal cells and reduction cells. The recently developed DARTS (Differentiable Architecture Search) is adopted to search for an effective network architecture that is built upon the two types of cells. All operations are 2D based in order to reduce the overall computation and search space. Experiments on the challenging NTU RGB+D and Kinectics datasets have verified that most of the networks developed to date for skeleton-based action recognition are likely not compact and efficient. The proposed method provides an approach to search for such a compact network that is able to achieve comparative or even better performance than the state-of-the-art methods.