cbx1/sam-vit-h 🖼️ → ❓

▶️ 349 runs 📅 Jan 2024 ⚙️ Cog 0.9.4 🔗 GitHub
image-embedding segment-anything

About

The Segment Anything Model (SAM) is a powerful and versatile image segmentation model. It leverages a "foundation model" approach, meaning it can be used for various segmentation tasks without needing to be specifically trained for each one.

Example Output

Output

{"image_embedding":"https://replicate.delivery/pbxt/DwFwpCxCKm6zEd93eyXIXsL8yhpNkTN8OYhOJ84CfUcE7iSSA/image_em.npy"}

Performance Metrics

2.88s Prediction Time
2.90s Total Time
Input Parameters
image (required) Type: string
Image to run embeddings on
Output Schema
image_embedding Type: stringFormat: uri
Image Embedding
Example Execution Logs 
loaded model and getting mask from image:  https://lh3.googleusercontent.com/5QxTBCxMUtA0NGvReyF34FXnJAuCuEH1mLBFSH-Ceo9BnkOVfN9FwZzzxAziuc2ntCHItWYrfpTRfsIuYJrBwv1SNLibnMcRSbyz3ABNMr2HpitgIONC=w600-l90-sg-rj-c0xffffff
starting
got embeddings
Elements in curr directory
__pycache__
.dockerignore
predict.py
cog.yaml
sam_vit_h_4b8939.pth
.cog
Version Details
Version ID
90b15ab8e2738ae5da8bdadb6e8644d083c0ddad3358867367bb038d038cad71
Version Created
January 31, 2024
Run on Replicate →