Video Camouflaged Object Detection (VCOD) focuses on identifying and segmenting objects
concealed within the background scenes. These camouflaged objects closely resemble their
surroundings by mimicking similar color patterns and textures, which poses significant challenges
compared to conventional detection tasks.
To address this problem, we have proposed a motion-enhanced approach that progressively
refines the detection results with multi-resolution search and motion-guided boosting. The video
frame is first screened under image level, and the inter-frame motions and background models are then corrected, considering all the video sequences. This method provides stable performance under popular VCOD datasets.