CODiff: One-Step Diffusion Model for Camouflaged Object Detection

Diffusion-based camouflaged object detection (COD) has recently shown great potential. In contrast to existing approaches that rely on multiple sample steps to refine the predicted masks, we propose CODiff, which reformulates the diffusion process to enable one-step mask prediction while maintaining competitive accuracy. Specifically, we first establish the theoretical feasibility of one-step sampling for COD. Based on this, we design a dedicated network for one-step inference with a global semantic guidance mechanism to guide the denoising process globally and hierarchical condition integration blocks to provide fine-grained structural semantics. In addition, we design a straight-forward regularization to learn better intermediate features by bridging the representation gap between the condition backbone and the diffusion model. Extensive experiments demonstrate that CODiff achieves state-of-the-art performance across multiple benchmarks, improving MAE by over 22\% on the challenging COD10K dataset. Code is available at https://github.com/KiiSooo/CODiff.

Existing works have explored diffusion models for camouflaged object detection (COD), such as identifying animals that seamlessly blend into complex backgrounds. Although these methods achieve strong performance, they typically require multiple inference steps, resulting in high computational cost and slow processing speed. We find that COD has the potential to be solved accurately with even a single inference step. By designing a framework tailored to this property, our method can effectively capture both complex structures and objects at different scales. Our work demonstrates the potential of few-step and even one-step diffusion models for discriminative vision tasks.