Learning Discriminative Prototypes with Dynamic Time Warping
This addresses the need for better temporal recognition and analysis in fields like video processing, though it appears incremental by building on existing DTW techniques.
The paper tackled the problem of learning discriminative prototypes for temporal data by proposing DP-DTW, which outperformed conventional DTW methods on time series classification benchmarks and achieved state-of-the-art results in weakly supervised action segmentation.
Dynamic Time Warping (DTW) is widely used for temporal data processing. However, existing methods can neither learn the discriminative prototypes of different classes nor exploit such prototypes for further analysis. We propose Discriminative Prototype DTW (DP-DTW), a novel method to learn class-specific discriminative prototypes for temporal recognition tasks. DP-DTW shows superior performance compared to conventional DTWs on time series classification benchmarks. Combined with end-to-end deep learning, DP-DTW can handle challenging weakly supervised action segmentation problems and achieves state of the art results on standard benchmarks. Moreover, detailed reasoning on the input video is enabled by the learned action prototypes. Specifically, an action-based video summarization can be obtained by aligning the input sequence with action prototypes.