InstructIR: High-Quality Image Restoration Following Human Instructions
This work addresses the challenge of flexible and high-quality image restoration for users by enabling natural language control, representing a novel benchmark in the field.
The paper tackles the problem of image restoration by introducing a method that uses human-written instructions to guide the model, achieving state-of-the-art results on tasks like denoising and deblurring with a +1dB improvement over previous all-in-one methods.
Image restoration is a fundamental problem that involves recovering a high-quality clean image from its degraded observation. All-In-One image restoration models can effectively restore images from various types and levels of degradation using degradation-specific information as prompts to guide the restoration model. In this work, we present the first approach that uses human-written instructions to guide the image restoration model. Given natural language prompts, our model can recover high-quality images from their degraded counterparts, considering multiple degradation types. Our method, InstructIR, achieves state-of-the-art results on several restoration tasks including image denoising, deraining, deblurring, dehazing, and (low-light) image enhancement. InstructIR improves +1dB over previous all-in-one restoration methods. Moreover, our dataset and results represent a novel benchmark for new research on text-guided image restoration and enhancement. Our code, datasets and models are available at: https://github.com/mv-lab/InstructIR