Align and Surpass Human Camouflaged Perception: Visual Refocus Reinforcement Fine-Tuning

TL;DR

This paper introduces a visual refocus reinforcement framework that enables multi-modal models to better identify camouflaged objects by mimicking human visual perception, achieving superior reasoning and detection performance.

Contribution

The paper proposes a novel visual refocus reinforcement learning approach that improves multi-modal models' ability to perceive camouflaged objects, surpassing existing fine-tuning methods.

Findings

Emergence of refocus visual phenomena with multiple reasoning tokens

Significant performance improvements in camouflaged object classification

Enhanced detection accuracy over baseline methods

Abstract

Current multi-modal models exhibit a notable misalignment with the human visual system when identifying objects that are visually assimilated into the background. Our observations reveal that these multi-modal models cannot distinguish concealed objects, demonstrating an inability to emulate human cognitive processes which effectively utilize foreground-background similarity principles for visual analysis. To analyze this hidden human-model visual thinking discrepancy, we build a visual system that mimicks human visual camouflaged perception to progressively and iteratively `refocus' visual concealed content. The refocus is a progressive guidance mechanism enabling models to logically localize objects in visual images through stepwise reasoning. The localization process of concealed objects requires hierarchical attention shifting with dynamic adjustment and refinement of prior cognitive knowledge. In this paper, we propose a visual refocus reinforcement framework via the policy optimization algorithm to encourage multi-modal models to think and refocus more before answering, and achieve excellent reasoning abilities to align and even surpass human camouflaged perception systems. Our extensive experiments on camouflaged perception successfully demonstrate the emergence of refocus visual phenomena, characterized by multiple reasoning tokens and dynamic adjustment of the detection box. Besides, experimental results on both camouflaged object classification and detection tasks exhibit significantly superior performance compared to Supervised Fine-Tuning (SFT) baselines.