MirrorNet: Bio-Inspired Camouflaged Object Segmentation
This work addresses the problem of detecting camouflaged objects in natural environments, which is challenging for both humans and machines, representing a domain-specific advancement.
The paper tackles camouflaged object segmentation by proposing MirrorNet, a bio-inspired network that uses dual streams for original and flipped images, achieving 89% accuracy on the CAMO dataset and outperforming state-of-the-art methods.
Camouflaged objects are generally difficult to be detected in their natural environment even for human beings. In this paper, we propose a novel bio-inspired network, named the MirrorNet, that leverages both instance segmentation and mirror stream for the camouflaged object segmentation. Differently from existing networks for segmentation, our proposed network possesses two segmentation streams: the main stream and the mirror stream corresponding with the original image and its flipped image, respectively. The output from the mirror stream is then fused into the main stream's result for the final camouflage map to boost up the segmentation accuracy. Extensive experiments conducted on the public CAMO dataset demonstrate the effectiveness of our proposed network. Our proposed method achieves 89% in accuracy, outperforming the state-of-the-arts. Project Page: https://sites.google.com/view/ltnghia/research/camo