Scale Adaptive Blind Deblurring

Scale Adaptive Blind Deblurring

-- A New Perspective on Robust Blind Deblurring

Advances in Neural Information Processing Systems (NIPS) Dec. 2014

Abstract: The presence of noise and small scale structures usually leads to large kernel estimation errors in blind image deblurring empirically, if not a total failure. We present a scale space perspective on blind deblurring algorithms, and introduce a cascaded scale space formulation for blind deblurring. This new formulation suggests a natural approach robust to noise and small scale structures through tying the estimation across multiple scales and balancing the contributions of different scales automatically by learning from data. The proposed formulation also allows to handle non-uniform blur with a straightforward extension. Experiments are conducted on both benchmark dataset and real-world images to validate the effectiveness of the proposed method. One surprising finding based on our approach is that blur kernel estimation is not necessarily best at the finest scale.

Uniform Deblurring for Noisy&Blurry Images

Blurry Tai et al. (CVPR'12) Zhong et al. (CVPR'13) Proposed Method

Blurry Tai et al. (CVPR'12)

Zhong et al. (CVPR'13) Proposed Method

Non-Uniform Deblurring

Blurry & Noisy Xu et al. (CVPR'13)

Zhong et al. (CVPR'13) Proposed Method

Related Publication and Software

Haichao Zhang and Jianchao Yang, Scale Adaptive Blind Deblurring, Advances in Neural Information and Processing Systems (NIPS) 2014 [PDF]