Correlation between Alignment-Uniformity and Performance of Dense Contrastive Representations
Jong Hak Moon (KAIST),* Wonjae Kim (NAVER AI Lab), Edward Choi (KAIST)
The 33rd British Machine Vision Conference
Abstract
Recently, dense contrastive learning has shown superior performance on dense prediction tasks compared to instance-level contrastive learning. Despite its supremacy, the properties of dense contrastive representations have not yet been carefully studied. Therefore, we analyze the theoretical ideas of dense contrastive learning using a standard CNN and straightforward feature matching scheme rather than propose a new complex method. Inspired by the analysis of the properties of instance-level contrastive representations through the lens of alignment and uniformity on the hypersphere, we employ and extend the same lens for the dense contrastive representations to analyze their underexplored properties. We discover the core principle in constructing a positive pair of dense features and empirically proved its validity. Also, we introduces a new scalar metric that summarizes the correlation between alignment-and-uniformity and downstream performance. Using this metric, we study various facets of densely learned contrastive representations such as how the correlation changes over single- and multi-object datasets or linear evaluation and dense prediction tasks.Video
Citation
@inproceedings{Moon_2022_BMVC,
author = {Jong Hak Moon and Wonjae Kim and Edward Choi},
title = {Correlation between Alignment-Uniformity and Performance of Dense Contrastive Representations},
booktitle = {33rd British Machine Vision Conference 2022, {BMVC} 2022, London, UK, November 21-24, 2022},
publisher = {{BMVA} Press},
year = {2022},
url = {https://bmvc2022.mpi-inf.mpg.de/0844.pdf}
}
Copyright © 2022
The British Machine Vision Conference is organised by