[All-in-one] 动态预训练 Dynamic Pre-training : 面向高效和可扩展的一体化图像恢复

Dynamic Pre-training: Towards Efficient and Scalable All-in-One Image Restoration

https://arxiv.org/pdf/2404.02154

GitHub

Abstract

All-in-one image restoration tackles different types of degradations with a unified model instead of having task-specific, non-generic models for each degradation. The requirement to tackle multiple degradations using the same model can lead to high-complexity designs with fixed configuration that lack the adaptability to more efficient alternatives. We propose DyNet, a dynamic family of networks designed in an encoder-decoder style for all-in-one image restoration tasks. Our DyNet can seamlessly switch between its bulkier and lightweight variants, thereby offering flexibility for efficient model deployment with a single round of training. This seamless switching is enabled by our weightssharing mechanism, forming the core of our architecture and facilitating the reuse of initialized module weights. Further, to establish robust weights initialization, we introduce a dynamic pre-training strategy that trains variants of the proposed DyNet concurrently, thereby achieving a 50% reduction in GPU hours. To tackle the unavailability of largescale dataset required in pre-training, we curate a high-quality, highresolution image dataset named Million-IRD having 2M image samples. We validate our DyNet for image denoising, deraining, and dehazing in all-in-one setting, achieving state-of-the-art results with 31.34% reduction in GFlops and a 56.75% reduction in parameters compared to baseline models.

动态预训练：迈向高效且可扩展的一站式图像恢复

一站式图像恢复通过统一的模型来处理不同类型的退化问题，而不是为每个退化问题设计任务特定的非通用模型。使用同一模型处