Semantic-aware representations for unsupervised Camouflaged Object Detection

Open Webpage

Zelin Lu, Xing Zhao, Liang Xie, Haoran Liang, Ronghua Liang

Published: 01 Jan 2025, Last Modified: 17 Apr 2025J. Vis. Commun. Image Represent. 2025EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Propose an innovative semantic-aware UCOD framework.•Propose a Multi-Scale Token loss for multi-scale detection.•Improve COD performance on four datasets without extra annotations.•Semantic inference avoids pseudo-label generation.