Bi-level Dynamic Learning for Jointly Multi-modality Image Fusion and Beyond
This addresses the challenge of integrating multi-modality scene perception tasks for intelligent vision systems, representing an incremental advancement by combining existing methods into a joint framework.
The paper tackles the problem of jointly optimizing multi-modality image fusion with downstream tasks like detection and segmentation, establishing a hierarchical deep model that bridges these tasks and demonstrating superior fused results and significant improvements in detection and segmentation over state-of-the-art methods.
Recently, multi-modality scene perception tasks, e.g., image fusion and scene understanding, have attracted widespread attention for intelligent vision systems. However, early efforts always consider boosting a single task unilaterally and neglecting others, seldom investigating their underlying connections for joint promotion. To overcome these limitations, we establish the hierarchical dual tasks-driven deep model to bridge these tasks. Concretely, we firstly construct an image fusion module to fuse complementary characteristics and cascade dual task-related modules, including a discriminator for visual effects and a semantic network for feature measurement. We provide a bi-level perspective to formulate image fusion and follow-up downstream tasks. To incorporate distinct task-related responses for image fusion, we consider image fusion as a primary goal and dual modules as learnable constraints. Furthermore, we develop an efficient first-order approximation to compute corresponding gradients and present dynamic weighted aggregation to balance the gradients for fusion learning. Extensive experiments demonstrate the superiority of our method, which not only produces visually pleasant fused results but also realizes significant promotion for detection and segmentation than the state-of-the-art approaches.