Learning Action Hierarchies via Hybrid Geometric Diffusion
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915708100673536 |
|---|---|
| author | Kaushik, Arjun Ramesh Ratha, Nalini K. Govindaraju, Venu |
| author_facet | Kaushik, Arjun Ramesh Ratha, Nalini K. Govindaraju, Venu |
| contents | Temporal action segmentation is a critical task in video understanding, where the goal is to assign action labels to each frame in a video. While recent advances leverage iterative refinement-based strategies, they fail to explicitly utilize the hierarchical nature of human actions. In this work, we propose HybridTAS - a novel framework that incorporates a hybrid of Euclidean and hyperbolic geometries into the denoising process of diffusion models to exploit the hierarchical structure of actions. Hyperbolic geometry naturally provides tree-like relationships between embeddings, enabling us to guide the action label denoising process in a coarse-to-fine manner: higher diffusion timesteps are influenced by abstract, high-level action categories (root nodes), while lower timesteps are refined using fine-grained action classes (leaf nodes). Extensive experiments on three benchmark datasets, GTEA, 50Salads, and Breakfast, demonstrate that our method achieves state-of-the-art performance, validating the effectiveness of hyperbolic-guided denoising for the temporal action segmentation task. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_01914 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Learning Action Hierarchies via Hybrid Geometric Diffusion Kaushik, Arjun Ramesh Ratha, Nalini K. Govindaraju, Venu Computer Vision and Pattern Recognition Temporal action segmentation is a critical task in video understanding, where the goal is to assign action labels to each frame in a video. While recent advances leverage iterative refinement-based strategies, they fail to explicitly utilize the hierarchical nature of human actions. In this work, we propose HybridTAS - a novel framework that incorporates a hybrid of Euclidean and hyperbolic geometries into the denoising process of diffusion models to exploit the hierarchical structure of actions. Hyperbolic geometry naturally provides tree-like relationships between embeddings, enabling us to guide the action label denoising process in a coarse-to-fine manner: higher diffusion timesteps are influenced by abstract, high-level action categories (root nodes), while lower timesteps are refined using fine-grained action classes (leaf nodes). Extensive experiments on three benchmark datasets, GTEA, 50Salads, and Breakfast, demonstrate that our method achieves state-of-the-art performance, validating the effectiveness of hyperbolic-guided denoising for the temporal action segmentation task. |
| title | Learning Action Hierarchies via Hybrid Geometric Diffusion |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.01914 |