“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

A method of video processing includes determining, for a conversion between a current block of a first picture of a video and a bitstream representation of the video, a motion candidate based on motion information from a second picture according to a rule. The rule specifies that a position from which the motion information is accessed is constrained to be within a specific subpicture of the second picture. The method also includes performing the conversion based on the determining.

Position restriction for inter coding mode

A method for video processing is provided to comprise: deriving, based on unclipped previously calculated luma motion vectors of multiple luma blocks, a chroma motion vector for a chroma block associated with a current video block of a video; and performing, based on the chroma motion vector, a conversion between the current video block and a bitstream representation of the video.

Motion vector derivation using higher bit-depth precision

A video processing method includes determining, for a conversion between a coding unit of a video and a bitstream representation of the video, use of a secondary transform with a reduced dimension (RST) tool to the coding unit based on a rule associated with one or more transform units in the coding unit and the coding unit. The RST tool comprises, during encoding, applying a forward secondary transform between a forward primary transform and a quantization step, or applying, during decoding, an inverse secondary transform between a de-quantization step and an inverse primary transform. Sizes of the forward secondary transform and the inverse secondary transform are less than a size of the coding unit. The method also includes performing the conversion based on the determining.

Block size dependent use of secondary transforms in coded video

A method of video processing is provided to comprising: determining, for a conversion between a current video block of a video and a coded representation of the video, a mode of operation of a generalized bi-prediction (GBi) coding tool in which a prediction of the current video block uses a final predictor corresponding to a non-uniformly weighted sum of predictors from two reference lists; and performing the conversion based on the determining, wherein a field in the coded representation corresponds to the mode of operation and a value of the field indicates whether the GBI coding tool is enabled or disabled for the current video block.

Generalized bi directional prediction mode in video processing

A method for visual media processing includes processing a bitstream representation for a conversion between a current video block in a video region of a visual media data and the bitstream representation. The bitstream representation is based on a rule that specifies that one or more syntax elements used in a luma-dependent chroma residual scaling step are selectively included in the bitstream representation. The luma-dependent chroma residual scaling step includes scaling chroma samples based on neighboring reconstructed luma samples during the conversion. The one or more syntax elements are usable to derive a chroma residual scaling factor used for the scaling.

Xu Jizheng

Papers

Position restriction for inter coding mode

Motion vector derivation using higher bit-depth precision

Block size dependent use of secondary transforms in coded video

Generalized bi directional prediction mode in video processing

Syntax element for chroma residual scaling