facebookresearch / facebookresearch/segment-anything

How to train SAM in batches?

Open
#445 5 comments 0 reactions 0 assignees View on GitHub
Dominant language
Jupyter Notebook
Stars
54.9k
Forks
6.4k
PR merge metrics
No merged PRs in 30d

Description

I have more than one target objects to segment in the image. Is it possible to feed batches into the model? It is possible to feed the batches into an image encoder. What about a prompt encoder? As I understand, the shape of the prompt batch should be BxNx4 for bounding boxes, where B is the batch size, N is the number of bounding boxes in image. I have checked the source code and prompt encoder expects a shape of Bx4. Am I missing something?

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.