OTAS: Unsupervised Boundary Detection for Object-Centric Temporal Action Segmentation
This work addresses the problem of temporal action segmentation for video analysis, offering an incremental improvement through a novel unsupervised method.
The paper tackles unsupervised boundary detection for temporal action segmentation by proposing OTAS, a framework using self-supervised global and local features, and reports that it outperforms the previous state-of-the-art method by 41% on average in F1 score and even exceeds human annotations in a user study.
Temporal action segmentation is typically achieved by discovering the dramatic variances in global visual descriptors. In this paper, we explore the merits of local features by proposing the unsupervised framework of Object-centric Temporal Action Segmentation (OTAS). Broadly speaking, OTAS consists of self-supervised global and local feature extraction modules as well as a boundary selection module that fuses the features and detects salient boundaries for action segmentation. As a second contribution, we discuss the pros and cons of existing frame-level and boundary-level evaluation metrics. Through extensive experiments, we find OTAS is superior to the previous state-of-the-art method by $41\%$ on average in terms of our recommended F1 score. Surprisingly, OTAS even outperforms the ground-truth human annotations in the user study. Moreover, OTAS is efficient enough to allow real-time inference.