Edge-Aware Deep Image Deblurring
This addresses the low-level vision problem of image deblurring for applications requiring sharp visual quality, but it is incremental as it builds on existing deep learning methods.
The paper tackles image deblurring by proposing a two-phase edge-aware deep network that enhances sharp edges, achieving promising results on standard benchmarks.
Image deblurring is a fundamental and challenging low-level vision problem. Previous vision research indicates that edge structure in natural scenes is one of the most important factors to estimate the abilities of human visual perception. In this paper, we resort to human visual demands of sharp edges and propose a two-phase edge-aware deep network to improve deep image deblurring. An edge detection convolutional subnet is designed in the first phase and a residual fully convolutional deblur subnet is then used for generating deblur results. The introduction of the edge-aware network enables our model with the specific capacity of enhancing images with sharp edges. We successfully apply our framework on standard benchmarks and promising results are achieved by our proposed deblur model.