VIFSS: View-Invariant and Figure Skating-Specific Pose Representation Learning for Temporal Action Segmentation
This work provides a practical solution for sports analytics in figure skating, enabling objective performance evaluation without expert knowledge, though it is incremental in improving domain-specific methods.
The paper tackled the problem of automating the recognition of figure skating jumps from videos by addressing limitations in temporal action segmentation, such as insufficient annotated data and lack of 3D and procedural modeling, and achieved over 92% F1@50 on element-level segmentation.
Understanding human actions from videos plays a critical role across various domains, including sports analytics. In figure skating, accurately recognizing the type and timing of jumps a skater performs is essential for objective performance evaluation. However, this task typically requires expert-level knowledge due to the fine-grained and complex nature of jump procedures. While recent approaches have attempted to automate this task using Temporal Action Segmentation (TAS), there are two major limitations to TAS for figure skating: the annotated data is insufficient, and existing methods do not account for the inherent three-dimensional aspects and procedural structure of jump actions. In this work, we propose a new TAS framework for figure skating jumps that explicitly incorporates both the three-dimensional nature and the semantic procedure of jump movements. First, we propose a novel View-Invariant, Figure Skating-Specific pose representation learning approach (VIFSS) that combines contrastive learning as pre-training and action classification as fine-tuning. For view-invariant contrastive pre-training, we construct FS-Jump3D, the first publicly available 3D pose dataset specialized for figure skating jumps. Second, we introduce a fine-grained annotation scheme that marks the ``entry (preparation)'' and ``landing'' phases, enabling TAS models to learn the procedural structure of jumps. Extensive experiments demonstrate the effectiveness of our framework. Our method achieves over 92% F1@50 on element-level TAS, which requires recognizing both jump types and rotation levels. Furthermore, we show that view-invariant contrastive pre-training is particularly effective when fine-tuning data is limited, highlighting the practicality of our approach in real-world scenarios.