… Given a set of severe motion blurred images, our bundle adjusted deblur NeRF (BAD-NeRF)
jointly learns the neural … It synthesizes novel images of higher quality than prior works. …

… To generate the blurred images, we applied a Gaussian kernel to each image, adjusting the
… We adjusted the set of object images to match the average power spectrum of the training …

… from blurry input. We model the blurring process by convolving a clean image using a blur
kernel … A novel deformable sparse kernel (DSK) module is proposed to model the blur kernel …

… Image blur is the chief cause of image degradation, and de-blurring turns to be a popular
research topic in image … causes of image blur, such as de-focus blur, Gaussian blur and motion …

… Besides, our method could easily adjust the blur effects of different exposure time during the
image capturing in the real world, by adaptively tuning translations of object and background…

… dataset in which blurred images have been classified using … neural networks are trained
by minimizing empirical risk for a finite number of image pairs, and parameters are adjusted to …

… myopes using similar methods of adjustment. Thresholds for … combined effects of optical and
neural limitations of the human … counteract the optical and neural attenuation of high spatial …

… in different image regions. We adopt multi-scale cross-correlation to handle blur and disparities
… DMENet [8] uses a deep neural network to only estimate a defocus map. Also, DPDNet+ …

… , and blur. In this work, we propose an Attention-guided Progressive Neural Texture Fusion
(… iris and brain are able to constantly adapt and adjust the responses to such strong stimulant …

… about specific parts of the input image to be focused on since blurry regions are not given, …
salient edges that require more attention can be adaptively adjusted to guide the learning of …