kahnchana/opl

Official repository for "Orthogonal Projection Loss" (ICCV'21)

42
/ 100
Emerging

This project offers a novel method to enhance the performance of deep learning models in image classification and related tasks. It takes standard image datasets with labels as input and produces more robust and accurate classification models by improving how the model separates different classes. Anyone working with image recognition, especially in fields requiring high accuracy or handling limited data, would find this beneficial.

129 stars. No commits in the last 6 months.

Use this if you are developing image classification models and need to improve accuracy, especially when dealing with domain generalization, few-shot learning, or robustness against adversarial attacks and noisy labels.

Not ideal if your primary goal is not image classification or if you are working with unsupervised learning tasks, as it's currently focused on supervised classification.

image-classification deep-learning-training computer-vision few-shot-learning model-robustness
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 16 / 25

How are scores calculated?

Stars

129

Forks

19

Language

Python

License

MIT

Last pushed

Nov 29, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/kahnchana/opl"

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