TokenMotion: Motion-Guided Vision Transformer for Video Camouflaged Object Detection Via Learnable Token Selection
It addresses the challenging problem of detecting camouflaged objects in videos for computer vision applications, representing an incremental advance with specific performance gains.
The paper tackles video camouflaged object detection by introducing TokenMotion, a transformer-based model that uses motion-guided features with learnable token selection, achieving state-of-the-art results including a 12.8% improvement in weighted F-measure on the MoCA-Mask dataset.
The area of Video Camouflaged Object Detection (VCOD) presents unique challenges in the field of computer vision due to texture similarities between target objects and their surroundings, as well as irregular motion patterns caused by both objects and camera movement. In this paper, we introduce TokenMotion (TMNet), which employs a transformer-based model to enhance VCOD by extracting motion-guided features using a learnable token selection. Evaluated on the challenging MoCA-Mask dataset, TMNet achieves state-of-the-art performance in VCOD. It outperforms the existing state-of-the-art method by a 12.8% improvement in weighted F-measure, an 8.4% enhancement in S-measure, and a 10.7% boost in mean IoU. The results demonstrate the benefits of utilizing motion-guided features via learnable token selection within a transformer-based framework to tackle the intricate task of VCOD.