Keypoint Detection
sapiens
English
rawalkhirodkar's picture
Update model card for Sapiens with architecture details
902ca43 verified
|
raw
history blame
1.41 kB
---
language: en
license: cc-by-nc-4.0
---
# Pose-Sapiens-1B-Torchscript
### Model Details
Sapiens is a family of vision transformers pretrained on 300 million human images at 1024 x 1024 image resolution. The pretrained models, when finetuned for human-centric vision tasks, generalize to in-the-wild conditions.
Sapiens-1B natively support 1K high-resolution inference. The resulting models exhibit remarkable generalization to in-the-wild data, even when labeled data is scarce or entirely synthetic.
- **Developed by:** Meta
- **Model type:** Vision Transformer
- **License:** Creative Commons Attribution-NonCommercial 4.0
- **Task:** pose
- **Format:** torchscript
- **File:** sapiens_1b_goliath_best_goliath_AP_640_torchscript.pt2
### Model Card
- **Image Size:** 1024 x 1024
- **Num Parameters:** 1.169 B
- **FLOPs:** 4.647 TFLOPs
- **Patch Size:** 16 x 16
- **Embedding Dimensions:** 1536
- **Num Layers:** 40
- **Num Heads:** 24
- **Feedforward Channels:** 6144
### More Sources
- **Repository:** [https://github.com/facebookresearch/sapiens](https://github.com/facebookresearch/sapiens)
- **Paper:** [https://arxiv.org/abs/2408.12569](https://arxiv.org/abs/2408.12569)
- **Demo:** [Sapiens Gradio Spaces](https://huggingface.co/collections/facebook/sapiens-66d22047daa6402d565cb2fc)
## Uses
Pose 1B model can be used for estimate 308 keypoints (body + face + hands + feet) on a single image.