Learning Spatial and Spatio-Temporal Pixel Aggregations for Image and Video Denoising

TL;DR

This paper introduces learned pixel aggregation networks for image and video denoising, which adaptively sample and average pixels to improve denoising quality, especially in dynamic scenes with large motion.

Contribution

The paper proposes novel deep neural networks for pixel aggregation that learn sampling and averaging strategies, advancing beyond hand-crafted methods for image and video denoising.

Findings

Outperforms state-of-the-art denoising methods on synthetic data

Effectively handles large motion in video denoising

Demonstrates improved results on real-world noisy images

Abstract

Existing denoising methods typically restore clear results by aggregating pixels from the noisy input. Instead of relying on hand-crafted aggregation schemes, we propose to explicitly learn this process with deep neural networks. We present a spatial pixel aggregation network and learn the pixel sampling and averaging strategies for image denoising. The proposed model naturally adapts to image structures and can effectively improve the denoised results. Furthermore, we develop a spatio-temporal pixel aggregation network for video denoising to efficiently sample pixels across the spatio-temporal space. Our method is able to solve the misalignment issues caused by large motion in dynamic scenes. In addition, we introduce a new regularization term for effectively training the proposed video denoising model. We present extensive analysis of the proposed method and demonstrate that our model performs favorably against the state-of-the-art image and video denoising approaches on both synthetic and real-world data.