Videomae Tennis Shottype
Fine-tuned VideoMAE-base (86M params) for 5-class tennis shot classification from 16-frame video clips.
Fuente del modelo
- Tipo
- Other
- Tamaño del archivo
- 329 MB
- Modelo base
- MCG-NJU/videomae-base
- Versión
- vRevision 23c812e
- Actualizado
- 24 oct 2026
- Descargas
- 259
- Fuente
- Hugging Face
Descripción de la fuente
VideoMAE Tennis Shot Type Classifier
Fine-tuned VideoMAE-base (86M params) for 5-class tennis shot classification from 16-frame video clips.
Classes
| Label | ID |
|---|---|
| Topspin | 0 |
| Slice | 1 |
| Volley | 2 |
| Flat | 3 |
| Smash | 4 |
Training Details
- Dataset: 2,436 annotated hit events from TenniSet (videos V006-V010)
- Input: 16 frames x 224x224, sampled around each hit timestamp
- Accuracy: 88-90% on held-out test split
- Hardware: Apple M4 Pro (48GB, MPS)
- Framework: HuggingFace Transformers
Usage
from transformers import VideoMAEImageProcessor, VideoMAEForVideoClassification
processor = VideoMAEImageProcessor.from_pretrained("vishnushenoy09/videomae-tennis-shottype")
model = VideoMAEForVideoClassification.from_pretrained("vishnushenoy09/videomae-tennis-shottype")
Credits
- VideoMAE by Zhan Tong, Yibing Song, Jue Wang, Limin Wang (MCG-NJU). Paper: VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training (NeurIPS 2022).
- TenniSet dataset by Kacper Wozniak. Used for training data annotation.
- Ball tracking and bounce detection adapted from yastrebksv/TrackNet and yastrebksv/TennisProject.
Part of
This model is part of the Sports Commentator from Video project — an automated tennis commentary pipeline using TrackNet, CatBoost, VideoMAE, Qwen2.5, RF-DETR, and F5-TTS.
Fuentes
1 fuente
Fuentes
1 fuenteVerificado 24 oct
Artefactos del modelo
1 artefacto
Artefactos del modelo
1 artefactoExtractos de fuentes
3 extractos
Extractos de fuentes
3 extractos--- license: cc-by-4.0 language: - en tags: - video-classification - videomae - tennis - sports - shot-type - fine-tuned base_model: MCG-NJU/videomae-base pipeline_tag: video-classification --- # VideoMAE Tennis Shot Type Classifier Fine-tuned [VideoMAE-base](https://huggingface.co/MCG-NJU/videomae-base) (86M params) for **5-class tennis shot classification** from 16-frame video clips. ## Classes | Label | ID | |-------|-----| | Topspin | 0 | | Slice | 1 | | Volley | 2 | | Flat | 3 | | Smash | 4 | ## Training Details - **Dataset:** 2,436 annotated hit events from TenniSet (videos V006-V010) - **Input:** 16 frames x 224x224, sampled around each hit timestamp - **Accuracy:** 88-90% on held-out test split - **Hardware:** Apple M4 Pro (48GB, MPS) - **Framework:** HuggingFace Transformers ## Usage ```python from transformers import VideoMAEImageProcessor, VideoMAEForVideoClassification processor = VideoMAEImageProcessor.from_pretrained("vishnushenoy09/videomae-tennis-shottype") model = VideoMAEForVideoClassification.from_pretrained("vishnushenoy09/videomae-tennis-shottype") ``` ## Credits - **VideoMAE** by Zhan Tong, Yibing Song, Jue Wang, Limin Wang (MCG-NJU). Paper: [VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training](https://arxiv.org/abs/2203.12602) (NeurIPS 2022). - **TenniSet** dataset by Kacper Wozniak. Used for training data annotation. - **Ball tracking and bounce detection** adapted from [yastrebksv/TrackNet](https://github.com/yastrebksv/TrackNet) and [yastrebksv/TennisProject](https://github.com/yastrebksv/TennisProject). ## Part of This model is part of the **Sports Commentator from Video** project — an automated tennis commentary pipeline using TrackNet, CatBoost, VideoMAE, Qwen2.5, RF-DETR, and F5-TTS.
Source context: 259 downloads · 0 likes · Pipeline video-classification · Repo vishnushenoy09/videomae-tennis-shottype
Source context: 16 downloads · 0 likes · Pipeline video-classification · Repo vishnushenoy09/videomae-tennis-shottype
