microsoft / microsoft/onnxruntime-inference-examples

Issue with segment-anything example

Open
#341 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
1.7k
Forks
414
Avg merge
1d 6h
Merged PRs (30d)
14

Description

@satyajandhyala Thank you so much for the segment-anything example, very helpful and amazing performance! One issue I've noticed, however, is that there is some issue with the scaling of the mask image:

image

I ran this same model in another app that I know works properly and it has the same scaling issue with these exported models. Is there something that needs to be changed in the way the model is exported to ONNX?

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by inspecting the segment-anything example and the ONNX model export path mentioned in the issue. Reproduce the mask scaling problem and compare the exported models with the known-working application. Done means identifying whether export or example processing causes the mismatch and correcting the model output scaling.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
computer-vision, machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.