TAPE: Task-Agnostic Prior Embedding for Image Restoration
This addresses the challenge of generalizing image priors across various degradation types for image restoration tasks, though it appears incremental as it builds on existing transformer and pre-training methods.
The paper tackles the problem of learning a generalized prior for natural image restoration by proposing TAPE, a two-stage approach that embeds a task-agnostic prior into a transformer, resulting in performance improvements of up to 1.45dB in PSNR and outperforming task-specific algorithms.
Learning a generalized prior for natural image restoration is an important yet challenging task. Early methods mostly involved handcrafted priors including normalized sparsity, l_0 gradients, dark channel priors, etc. Recently, deep neural networks have been used to learn various image priors but do not guarantee to generalize. In this paper, we propose a novel approach that embeds a task-agnostic prior into a transformer. Our approach, named Task-Agnostic Prior Embedding (TAPE), consists of two stages, namely, task-agnostic pre-training and task-specific fine-tuning, where the first stage embeds prior knowledge about natural images into the transformer and the second stage extracts the knowledge to assist downstream image restoration. Experiments on various types of degradation validate the effectiveness of TAPE. The image restoration performance in terms of PSNR is improved by as much as 1.45dB and even outperforms task-specific algorithms. More importantly, TAPE shows the ability of disentangling generalized image priors from degraded images, which enjoys favorable transfer ability to unknown downstream tasks.