facebook/sapiens-depth-0.3b-bfloat16

Depth-Sapiens-0.3B-Bfloat16

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-0.3B 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: depth
• Format: bfloat16
• File: sapiens_0.3b_render_people_epoch_100_bfloat16.pt2

Model Card

• Image Size: 1024 x 768 (H x W)
• Num Parameters: 0.336 B
• FLOPs: 1.242 TFLOPs
• Patch Size: 16 x 16
• Embedding Dimensions: 1024
• Num Layers: 24
• Num Heads: 16
• Feedforward Channels: 4096

More Resources

• Repository: https://github.com/facebookresearch/sapiens
• Paper: https://arxiv.org/abs/2408.12569
• Demo: https://huggingface.co/spaces/facebook/sapiens-depth
• Project Page: https://about.meta.com/realitylabs/codecavatars/sapiens
• Additional Results: https://rawalkhirodkar.github.io/sapiens
• HuggingFace Collection: https://huggingface.co/collections/facebook/sapiens-66d22047daa6402d565cb2fc

Uses

Depth 0.3B model can be used to estimate relative depth on human images.