ananthu-aniraj/masking_strategies_bias_removal
Masking Strategies for Background Bias Removal in Computer Vision Models (ICCVW OODCV 2023 paper)
This project helps scientists, researchers, or anyone working with fine-grained image classification avoid common pitfalls where AI models mistakenly associate an object with its background rather than its true characteristics. It takes your existing image datasets, like those used for identifying bird species, and outputs more robust classification models. The ideal user is a data scientist or researcher building and evaluating AI models for detailed image analysis.
No commits in the last 6 months.
Use this if your image classification models are struggling with accuracy because they're being fooled by image backgrounds, especially when encountering images with unfamiliar or 'out-of-distribution' backgrounds.
Not ideal if you are working on object detection or general image recognition tasks where background context is often helpful, or if you don't have labeled segmentation masks for your dataset.
Stars
16
Forks
1
Language
Python
License
MIT
Category
Last pushed
Jul 03, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ananthu-aniraj/masking_strategies_bias_removal"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
Westlake-AI/openmixup
CAIRI Supervised, Semi- and Self-Supervised Visual Representation Learning Toolbox and Benchmark
YU1ut/MixMatch-pytorch
Code for "MixMatch - A Holistic Approach to Semi-Supervised Learning"
kamata1729/QATM_pytorch
Pytorch Implementation of QATM:Quality-Aware Template Matching For Deep Learning
nttcslab/msm-mae
Masked Spectrogram Modeling using Masked Autoencoders for Learning General-purpose Audio Representations
rgeirhos/generalisation-humans-DNNs
Data, code & materials from the paper "Generalisation in humans and deep neural networks" (NeurIPS 2018)