Realistic Blur Synthesis for Learning Image Deblurring

17 Feb 2022  ·  Jaesung Rim, Geonung Kim, Jungeon Kim, Junyong Lee, Seungyong Lee, Sunghyun Cho ·

Training learning-based deblurring methods demands a tremendous amount of blurred and sharp image pairs. Unfortunately, existing synthetic datasets are not realistic enough, and deblurring models trained on them cannot handle real blurred images effectively. While real datasets have recently been proposed, they provide limited diversity of scenes and camera settings, and capturing real datasets for diverse settings is still challenging. To resolve this, this paper analyzes various factors that introduce differences between real and synthetic blurred images. To this end, we present RSBlur, a novel dataset with real blurred images and the corresponding sharp image sequences to enable a detailed analysis of the difference between real and synthetic blur. With the dataset, we reveal the effects of different factors in the blur generation process. Based on the analysis, we also present a novel blur synthesis pipeline to synthesize more realistic blur. We show that our synthesis pipeline can improve the deblurring performance on real blurred images.

PDF Abstract


Introduced in the Paper:


Used in the Paper:

GoPro Real Blur Dataset
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Deblurring RSBlur (trained on synthetic) MIMO-UNet + Realistic blur Average PSNR 32.08 # 1
Deblurring RSBlur (trained on synthetic) SRN-Deblur + Realistic blur Average PSNR 32.06 # 2


No methods listed for this paper. Add relevant methods here