Image resolution

Lidar
https://paperswithcode.com/paper/a-dataset-for-semantic-segmentation-of-point

Lidar

shift invariant
https://paperswithcode.com/paper/190411486

https://richzhang.github.io/antialiased-cnns/

https://github.com/hendrycks/robustness Modern convolutional networks are not shift-invariant, as small input shifts or translations can cause drastic changes in the output. Commonly used downsampling methods, such as max-pooling, strided-convolution, and average-pooling, ignore the sampling theorem. The well-known signal processing fix is anti-aliasing by low-pass filtering before downsampling. However, simply inserting this module into deep networks degrades performance; as a result, it is seldomly used today. We show that when integrated correctly, it is compatible with existing architectural components, such as max-pooling and strided-convolution. We observe \textit{increased accuracy} in ImageNet classification, across several commonly-used architectures, such as ResNet, DenseNet, and MobileNet, indicating effective regularization. Furthermore, we observe \textit{better generalization}, in terms of stability and robustness to input corruptions. Our results demonstrate that this classical signal processing technique has been undeservingly overlooked in modern deep networks. Code and anti-aliased versions of popular networks are available at https://richzhang.github.io/antialiased-cnns/

image restore
https://github.com/cszn/DPSR

https://arxiv.org/abs/1903.12529Deep Plug-and-Play Super-Resolution for Arbitrary Blur Kernels (CVPR, 2019). While deep neural networks (DNN) based single image super-resolution (SISR) methods are rapidly gaining popularity, they are mainly designed for the widely-used bicubic degradation, and there still remains the fundamental challenge for them to super-resolve low-resolution (LR) image with arbitrary blur kernels. In the meanwhile, plug-and-play image restoration has been recognized with high flexibility due to its modular structure for easy plug-in of denoiser priors. In this paper, we propose a principled formulation and framework by extending bicubic degradation based deep SISR with the help of plug-and-play framework to handle LR images with arbitrary blur kernels. Specifically, we design a new SISR degradation model so as to take advantage of existing blind deblurring methods for blur kernel estimation. To optimize the new degradation induced energy function, we then derive a plug-and-play algorithm via variable splitting technique, which allows us to plug any super-resolver prior rather than the denoiser prior as a modular part. Quantitative and qualitative evaluations on synthetic and real LR images demonstrate that the proposed deep plug-and-play super-resolution framework is flexible and effective to deal with blurry LR images. https://arxiv.org/abs/1903.12529

super resolution
https://www.youtube.com/watch?v=HvH0b9K_Iro site http://igl.ethz.ch/projects/prosr/ and github page https://github.com/fperazzi/proSR

https://github.com/aiff22/DPED by http://people.ee.ethz.ch/~ihnatova/

https://www.youtube.com/watch?v=WovbLx8C0yA Google super resolution at https://github.com/JalaliLabUCLA/Jalali-Lab-Implementation-of-RAISR

noise removal
https://github.com/DmitryUlyanov/deep-image-prior from Neural papers with code

links
Neural papers with code

Semantic segmentation