Motion Deblurring in the Wild
This work addresses motion deblurring for images taken in the wild, which is crucial for improving photo quality in uncontrolled environments, representing a strong specific gain rather than a foundational advancement.
The paper tackles the challenging problem of motion deblurring in real-world images by proposing a three-stage convolutional network architecture and creating a realistic dataset from high-frame-rate videos, achieving state-of-the-art performance and effectively handling occlusions.
The task of image deblurring is a very ill-posed problem as both the image and the blur are unknown. Moreover, when pictures are taken in the wild, this task becomes even more challenging due to the blur varying spatially and the occlusions between the object. Due to the complexity of the general image model we propose a novel convolutional network architecture which directly generates the sharp image.This network is built in three stages, and exploits the benefits of pyramid schemes often used in blind deconvolution. One of the main difficulties in training such a network is to design a suitable dataset. While useful data can be obtained by synthetically blurring a collection of images, more realistic data must be collected in the wild. To obtain such data we use a high frame rate video camera and keep one frame as the sharp image and frame average as the corresponding blurred image. We show that this realistic dataset is key in achieving state-of-the-art performance and dealing with occlusions.