facebookresearch / facebookresearch/sam3

sam3 supports image input as a prompt, and a box needs to be provided for input. Can we develop an interface where the input is changed to a numpy format of an image? I think this would be more flexible.

Open
#490 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
11.7k
Forks
1.8k
PR merge metrics
No merged PRs in 30d

Description

sam3 supports image input as a prompt, and a box needs to be provided for input. Can we develop an interface where the input is changed to a numpy format of an image? I think this would be more flexible.

Contributor guide

Open the contributing guide

Research direction

Start by inspecting the SAM 3 image-prompt inference interface and the example notebooks to find where image and box inputs are defined. Determine the intended NumPy image shape and how it should coexist with the required box; done means the interface accepts a NumPy image input and the documented example works without changing existing prompt behavior.

Written by the indexing model from the issue text.

Assessment

Tech stack
numpy, python
Domain
computer-vision, machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.