“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

The PASCAL Visual Object Classes Challenge

The compression capability of several generations of video coding standards is compared by means of peak signal-to-noise ratio (PSNR) and subjective testing results. A unified approach is applied to the analysis of designs, including H.262/MPEG-2 Video, H.263, MPEG-4 Visual, H.264/MPEG-4 Advanced Video Coding (AVC), and High Efficiency Video Coding (HEVC). The results of subjective tests for WVGA and HD sequences indicate that HEVC encoders can achieve equivalent subjective reproduction quality as encoders that conform to H.264/MPEG-4 AVC when using approximately 50% less bit rate on average. The HEVC design is shown to be especially effective for low bit rates, high-resolution video content, and low-delay communication applications. The measured subjective improvement somewhat exceeds the improvement measured by the PSNR metric.

http://amalia.img.lx.it.pt/~fp/cav/Additional_material/HEVC-Performance.pdf

Comparison of the Coding Efficiency of Video Coding Standards—Including High Efficiency Video Coding (HEVC)

We present DeblurGAN, an end-to-end learned method for motion deblurring. The learning is based on a conditional GAN and the content loss. DeblurGAN achieves state-of-the art performance both in the structural similarity measure and visual appearance. The quality of the deblurring model is also evaluated in a novel way on a real-world problem - object detection on (de-)blurred images. The method is 5 times faster than the closest competitor - Deep-Deblur [25]. We also introduce a novel method for generating synthetic motion blurred images from sharp ones, allowing realistic dataset augmentation. The model, code and the dataset are available at https://github.com/KupynOrest/DeblurGAN

/pdf/deblurgan-blind-motion-deblurring-using-conditional-150wx51y9w.pdf

DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks

We present a comprehensive study and evaluation of existing single-image dehazing algorithms, using a new large-scale benchmark consisting of both synthetic and real-world hazy images, called REalistic Single-Image DEhazing (RESIDE). RESIDE highlights diverse data sources and image contents, and is divided into five subsets, each serving different training or evaluation purposes. We further provide a rich variety of criteria for dehazing algorithm evaluation, ranging from full-reference metrics to no-reference metrics and to subjective evaluation, and the novel task-driven evaluation. Experiments on RESIDE shed light on the comparisons and limitations of the state-of-the-art dehazing algorithms, and suggest promising future directions.

Benchmarking Single-Image Dehazing and Beyond

Multi-HMVP for affine is described. In an exemplary aspect, a method for video processing includes deriving, for a conversion between a current block of video and a bitstream representation of the current block, an affine model for the current block based on affine related information stored in at least one history motion vector prediction (HMVP) buffer of multiple HMVP buffers, the multiple HMVP buffers being used to store affine related information of blocks coded prior the current block in different categories; and performing the conversion by using the derived affine model.

Multi-hmvp for affine

A video processing method includes determining, for a conversion between a video block of a video and a coded representation of the video, factors of a scaling tool based on a coding mode of the video block; and performing the conversion using the scaling tool, wherein the use of the scaling tool comprises: scaling at least some coefficients representing the video block during encoding or descaling at least some coefficients from the coded representation during decoding.

Usage of default and user-defined scaling matrices

Methods, systems, and devices for high-precision transform and quantization for image and video coding are described. A example method of video processing includes determining, for a conversion between a video comprising a current block and a bitstream representation of the video, that the conversion comprises an application of a transform to the current block, and performing, based on the determining, the conversion, wherein a bit-shifting operation of a scaling process for transform coefficients associated with the transform is based on whether the current block is coded with a block-based differential pulse code modulation (BDPCM) mode.

High precision transform and quantization for image and video coding

Devices, systems and methods for digital video coding, which includes deriving RMVF-based motion vector information, are described. An exemplary method for video processing includes deriving at least one motion model for a current video block, based on motion information of at least one non-adjacent spatial neighboring block or at least one temporal neighboring block of the current video block; deriving motion information for the current video block or at least one sub-block of the current video block based on the at least one motion model; and performing a conversion for the current video block based on the derived motion information.

Regression-based motion vector field based sub-block motion vector derivation

A method for video processing is described, the method comprises: making a decision, for a conversion between a current video block of a video picture of a video and a coded representation of the video, regarding a size of a reference region of the video picture from which reference samples are used for predicting the current video block, based on a size of a virtual pipeline data unit (VPDU), a size of a coding tree block (CTB), or a size of a coding tree unit (CTU) (812), and, performing the conversion based on the decision (814).

Xu Jizheng

Papers

Multi-hmvp for affine

Usage of default and user-defined scaling matrices

High precision transform and quantization for image and video coding

Regression-based motion vector field based sub-block motion vector derivation

Size restriction for intra-block copy virtual buffer