Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

How to get confidence score from images matching #88

Open
KadriMufti opened this issue Oct 9, 2024 · 2 comments
Open

How to get confidence score from images matching #88

KadriMufti opened this issue Oct 9, 2024 · 2 comments
Labels
question Further information is requested

Comments

@KadriMufti
Copy link

Checklist

  • [X ] I've searched the project's [issues]

❓ Question

I am attempting to use this repo and the webUI to match between photographs of people. How can I get a single confidence score or matching score? I am also unsure if this repo is useful for profile photo matching.

📎 Additional context

The Match Info and Matches Statistics sections of the webUI do not provide a final single score that I need.
E.g.
{"num_raw_matches": 217, "num_ransac_matches": 190 }

@KadriMufti KadriMufti added the question Further information is requested label Oct 9, 2024
@Avinash24R
Copy link

Here is a piece of code from my small project.
#Code

predictions = model.predict(testX)
y_pred = np.argmax(predictions, axis=1)  # Convert predictions to class labels

# Convert testY from one-hot encoded to class labels (integers)
y_true = np.argmax(testY, axis=1)

print(classification_report(y_true, y_pred, target_names=le.classes_, digits = 5))

@Vincentqyw
Copy link
Owner

From the information you provided, you can calculate the average match score from feature points, such as those provided by SuperGlue or other nn matches.

matches, confid = pred["matches0"], pred["matching_scores0"]

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
question Further information is requested
Projects
None yet
Development

No branches or pull requests

3 participants