To Blend In, First Decouple: Rethinking Camouflage Image Generation via Context-Decoupled Representations

2026-07-20

AI Summary

Existing camouflage image generation methods often suffer from appearance inconsistency and background artifacts due to the tight coupling between object and background representations. To address this limitation, this work proposes CamoDreamer, which introduces a novel context-decoupled generation paradigm. By leveraging contrastive-aware conditional guidance, the method separately models object and background features, then coherently integrates high-frequency textures and low-frequency structures in the latent space through a context-decoupled assimilation flow and a frequency-adaptive fusion mechanism. This approach substantially enhances both camouflage consistency and controllability while maintaining a lightweight and efficient architecture, outperforming current state-of-the-art methods across multiple quantitative metrics.

Abstract

Camouflage image generation (CIG) focuses on generating visually concealed objects that seamlessly blend into their backgrounds. Existing methods typically follow either background-guided paradigms that adapt object appearance via style transfer, or foreground-guided strategies that outpaint surrounding regions conditioned on object features. However, they still suffer from appearance discrepancy and background artifacts. We attribute these limitations to cross-context representation leakage, where object and background cues are entangled in a coupled conditional space, resulting in ambiguous control and degraded camouflage fidelity. To tackle this, we propose a new context-decoupled generative paradigm, termed CamoDreamer, which aims to isolate contextual conditional guidance and explicitly decouple latent camouflage features into coordinated object and background control streams. First, a Contrast-aware Contextual Bridge is designed to model cross-context discrepancies and construct contrast-aware dual conditional guidance. Second, Context-Decoupled Assimilation Streams are employed to separate generative interactions conditioned on the dual guidance, while facilitating background rendering with target-aware cues in the latent space. Finally, a Frequency-Adaptive Contextual Blend module integrates complementary high-frequency textures and low-frequency structures from decoupled features to improve holistic coherence. Extensive experiments demonstrate that CamoDreamer consistently outperforms existing methods with a substantial margin, while maintaining a relatively lightweight design.

Problem

• camouflage image generation
• appearance discrepancy
• background artifacts
• cross-context representation leakage
• context decoupling

Innovation

• context-decoupled representation
• camouflage image generation
• contrast-aware guidance
• latent feature disentanglement
• frequency-adaptive blending