ViCE: Improving Dense Representation Learning by Superpixelization and Contrasting Cluster Assignment

Recent self-supervised models have demonstrated equal or better performance than supervised methods, opening for AI systems to learn visual representations from practically unlimited data. However, these methods are typically classification-based and thus ineffective for learning high-resolution fea...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2022-10
Main Authors: Karlsson, Robin, Hayashi, Tomoki, Fujii, Keisuke, Carballo, Alexander, Ohtani, Kento, Takeda, Kazuya
Format: Article
Language:English
Subjects:
Image resolution
Pixels
Representations
Stochastic processes
Supervised learning
Visual observation
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Recent self-supervised models have demonstrated equal or better performance than supervised methods, opening for AI systems to learn visual representations from practically unlimited data. However, these methods are typically classification-based and thus ineffective for learning high-resolution feature maps that preserve precise spatial information. This work introduces superpixels to improve self-supervised learning of dense semantically rich visual concept embeddings. Decomposing images into a small set of visually coherent regions reduces the computational complexity by \(\mathcal{O}(1000)\) while preserving detail. We experimentally show that contrasting over regions improves the effectiveness of contrastive learning methods, extends their applicability to high-resolution images, improves overclustering performance, superpixels are better than grids, and regional masking improves performance. The expressiveness of our dense embeddings is demonstrated by improving the SOTA unsupervised semantic segmentation benchmark on Cityscapes, and for convolutional models on COCO.
ISSN:2331-8422