FineGym: A Hierarchical Video Dataset for Fine-grained Action Understanding
| Autor: | Yue Zhao, Bo Dai, Dahua Lin, Dian Shao |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2020 |
| Předmět: |
FOS: Computer and information sciences
Class (computer programming) Sequence Parsing Computer science business.industry media_common.quotation_subject Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition 02 engineering and technology 010501 environmental sciences computer.software_genre Semantics 01 natural sciences Action (philosophy) 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Quality (business) Artificial intelligence Set (psychology) business computer Natural language processing 0105 earth and related environmental sciences media_common |
| Zdroj: | CVPR |
| Popis: | On public benchmarks, current action recognition techniques have achieved great success. However, when used in real-world applications, e.g. sport analysis, which requires the capability of parsing an activity into phases and differentiating between subtly different actions, their performances remain far from being satisfactory. To take action recognition to a new level, we develop FineGym, a new dataset built on top of gymnastic videos. Compared to existing action recognition datasets, FineGym is distinguished in richness, quality, and diversity. In particular, it provides temporal annotations at both action and sub-action levels with a three-level semantic hierarchy. For example, a "balance beam" event will be annotated as a sequence of elementary sub-actions derived from five sets: "leap-jump-hop", "beam-turns", "flight-salto", "flight-handspring", and "dismount", where the sub-action in each set will be further annotated with finely defined class labels. This new level of granularity presents significant challenges for action recognition, e.g. how to parse the temporal structures from a coherent action, and how to distinguish between subtly different action classes. We systematically investigate representative methods on this dataset and obtain a number of interesting findings. We hope this dataset could advance research towards action understanding. CVPR 2020 Oral (3 strong accepts); Project page: https://sdolivia.github.io/FineGym/ |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|