Dynamic Slimmable Denoising Network

TL;DR

The paper introduces DDS-Net, a dynamic, channel-adjustable neural network for image denoising that adapts to different noisy images at test time, reducing computational costs while maintaining high quality.

Contribution

The work proposes a novel dynamic slimmable network with a dynamic gate and a three-stage training scheme for efficient, adaptive image denoising, outperforming static models.

Findings

Outperforms state-of-the-art static denoising networks.

Achieves comparable denoising quality with less computation.

Demonstrates effective dynamic adaptation to diverse noisy images.

Abstract

Recently, tremendous human-designed and automatically searched neural networks have been applied to image denoising. However, previous works intend to handle all noisy images in a pre-defined static network architecture, which inevitably leads to high computational complexity for good denoising quality. Here, we present dynamic slimmable denoising network (DDS-Net), a general method to achieve good denoising quality with less computational complexity, via dynamically adjusting the channel configurations of networks at test time with respect to different noisy images. Our DDS-Net is empowered with the ability of dynamic inference by a dynamic gate, which can predictively adjust the channel configuration of networks with negligible extra computation cost. To ensure the performance of each candidate sub-network and the fairness of the dynamic gate, we propose a three-stage optimization scheme. In the first stage, we train a weight-shared slimmable super network. In the second stage, we evaluate the trained slimmable super network in an iterative way and progressively tailor the channel numbers of each layer with minimal denoising quality drop. By a single pass, we can obtain several sub-networks with good performance under different channel configurations. In the last stage, we identify easy and hard samples in an online way and train a dynamic gate to predictively select the corresponding sub-network with respect to different noisy images. Extensive experiments demonstrate our DDS-Net consistently outperforms the state-of-the-art individually trained static denoising networks.