Skip to content
Jimmy Joseph
Go back

Completely Self-Supervised Crowd Counting via Distribution Matching

Published in the European Conference on Computer Vision (ECCV), 2022.

This work explores crowd counting through a completely self-supervised distribution-matching approach, contributing to research on learning from visual data without conventional count annotations.

Publication

Read the publication


Share this post:

Previous Post
The Protocol Genome: A Self-Supervised Learning Framework from DICOM Headers