PCA Whitening

Computer VisionIntroduced 20006 papers

Description

PCA Whitening is a processing step for image based data that makes input less redundant. Adjacent pixel or feature values can be highly correlated, and whitening through the use of PCA reduces this degree of correlation.

Image Source: Wikipedia

Papers Using This Method

Towards scientific discovery with dictionary learning: Extracting biological concepts from microscopy foundation models2024-12-20Zipfian Whitening2024-11-01Efficient Probabilistic Modeling of Crystallization at Mesoscopic Scale2024-05-26MultiGrain: a unified image embedding for classes and instances2019-02-14Attention-based Pyramid Aggregation Network for Visual Place Recognition2018-08-01Decorrelated Batch Normalization2018-04-23