Conditional Controllable Image Fusion

TL;DR

This work proposes a conditional controllable fusion (CCF) framework for general image fusion tasks without specific training, and first injects the specific constraints into the pre-trained DDPM as adaptive fusion conditions.

Abstract

Image fusion aims to integrate complementary information from multiple input images acquired through various sources to synthesize a new fused image. Existing methods usually employ distinct constraint designs tailored to specific scenes, forming fixed fusion paradigms. However, this data-driven fusion approach is challenging to deploy in varying scenarios, especially in rapidly changing environments. To address this issue, we propose a conditional controllable fusion (CCF) framework for general image fusion tasks without specific training. Due to the dynamic differences of different samples, our CCF employs specific fusion constraints for each individual in practice. Given the powerful generative capabilities of the denoising diffusion model, we first inject the specific constraints into the pre-trained DDPM as adaptive fusion conditions. The appropriate conditions are dynamically selected to ensure the fusion process remains responsive to the specific requirements in each reverse diffusion stage. Thus, CCF enables conditionally calibrating the fused images step by step. Extensive experiments validate our effectiveness in general fusion tasks across diverse scenarios against the competing methods without additional training.

Figures (12)

• Figure 1: The conditions selection statistics during the sampling process of the LLVIP dataset. The distinct process of sampling has different favor of the conditions. The crucial role that diverse conditions play in controlling various image generation processes. Throughout the diffusion sampling, different conditions are dynamically selected to best suit the generation requirement at each stage.
• Figure 2: Illustrates the pipeline of the proposed CCF. The framework comprises two components: a sampling process utilizing a pre-trained DDPM and a condition bank with SCS.
• Figure 3: Qualitative comparisons of our CCF and the competing methods on VIF fusion task of LLVIP dataset.
• Figure 4: Qualitative comparisons of various methods in MFF task from MFFW dataset.
• Figure 5: The visualization of the w/o and with task-specific conditions.