Lumina-OmniLV: A Unified Multimodal Framework for General Low-Level Vision
This work addresses the need for robust and generalizable systems in low-level vision, though it is incremental as it builds on existing methods like Diffusion Transformers.
The paper tackled the problem of developing a unified multimodal framework for over 100 low-level vision tasks, achieving optimal performance at 1K resolution with fine-grained detail preservation.
We present Lunima-OmniLV (abbreviated as OmniLV), a universal multimodal multi-task framework for low-level vision that addresses over 100 sub-tasks across four major categories: image restoration, image enhancement, weak-semantic dense prediction, and stylization. OmniLV leverages both textual and visual prompts to offer flexible and user-friendly interactions. Built on Diffusion Transformer (DiT)-based generative priors, our framework supports arbitrary resolutions -- achieving optimal performance at 1K resolution -- while preserving fine-grained details and high fidelity. Through extensive experiments, we demonstrate that separately encoding text and visual instructions, combined with co-training using shallow feature control, is essential to mitigate task ambiguity and enhance multi-task generalization. Our findings also reveal that integrating high-level generative tasks into low-level vision models can compromise detail-sensitive restoration. These insights pave the way for more robust and generalizable low-level vision systems.