mala-lab/NegPrompt

The official implementation of CVPR 24' Paper "Learning Transferable Negative Prompts for Out-of-Distribution Detection"

36
/ 100
Emerging

This project helps computer vision practitioners accurately identify images that fall outside of expected categories, even if those unexpected images are new and haven't been seen before. It takes a dataset of known image categories and learns to recognize what doesn't belong, providing a robust system for flagging out-of-distribution content. This is for machine learning engineers and researchers working with image classification systems who need to prevent misclassification of novel or anomalous data.

No commits in the last 6 months.

Use this if you need to build a highly reliable image classification system that can effectively detect and reject images from categories not present in your training data, even when those 'out-of-distribution' categories are completely new.

Not ideal if your primary goal is to simply classify images into pre-defined categories without needing to identify and flag novel, unseeable content.

anomaly-detection image-classification computer-vision machine-learning-operations open-set-recognition
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

61

Forks

7

Language

Python

License

MIT

Last pushed

Apr 08, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/prompt-engineering/mala-lab/NegPrompt"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.