Search

Deep learning has achieved great success in the past decade, especially in the fields of computer vision and image processing. After witnessing such success, video coding experts are motivated to consider whether deep learning can also benefit video coding, and if so, they seek to discover why and how. Indeed, a number of research studies have been conducted to explore deep learning for image and video coding, which has been an active and fast-growing research area especially since the year 2015. These prior arts can be divided into two categories: new coding schemes that are built solely upon deep networks (deep schemes), and deep network-based coding tools that are embedded into traditional coding schemes (deep tools). Moreover, in the responses to the joint call for proposals on video compression with capability beyond High Efficiency Video Coding (HEVC), a number of deep tools have been proposed, and some of them are further studied for the upcoming Versatile Video Coding (VVC). In this paper, we summarize the ongoing efforts in the Joint Video Experts Team about the proposed deep tools, and we discuss several promising tools in much detail, including neural network-based intra prediction, convolutional neural network (CNN) based in-loop filtering, and CNN-based block-adaptive-resolution coding. A series of experimental results are provided to demonstrate the capability of these tools in achieving higher compression efficiency than the VVC or HEVC anchor. These results shed light on the promising direction of deep learning-based future video coding, towards which a lot of open problems call for further study. [ABSTRACT FROM AUTHOR]

Recently, there has been a significant progress in the research and development of the high dynamic range (HDR) video technology and the state-of-the-art video pipelines are able to offer a higher bit depth support to capture, store, encode, and display HDR video content. In this paper, we introduce a novel HDR video compression algorithm, which uses a perceptually uniform color opponent space, a novel perceptual transfer function to encode the dynamic range of the scene, and a novel error minimization scheme for accurate chroma reproduction. The proposed algorithm was objectively and subjectively evaluated against four state-of-the-art algorithms. The objective evaluation was conducted across a set of 39 HDR video sequences, using the latest x265 10-bit video codec along with several perceptual and structural quality assessment metrics at 11 different quality levels. Furthermore, a rating-based subjective evaluation ($n=40$) was conducted with six sequences at two different output bitrates. Results suggest that the proposed algorithm exhibits the lowest coding error amongst the five algorithms evaluated. Additionally, the rate–distortion characteristics suggest that the proposed algorithm outperforms the existing state-of-the-art at bitrates ≥ 0.4 bits/pixel. [ABSTRACT FROM AUTHOR]

A residue number system (RNS) has an inherent parallel structure that can be utilized for improving computer hardware systems. An RNS represents large integer numbers as a smaller integer set, or residues of a modulo set, without carry propagation between them. Hence mathematical operations, such as addition or subtraction, can be performed on the residues independently. This paper proposes an RNS implementation of motion estimation for the latest video coding standard known as high-efficiency video coding (HEVC) or H.265. Since motion estimation is the most computationally intensive task in video coding, several simplified algorithms are proposed for mitigating the problem, but the majority of them result in a worsening peak signal-to-noise ratio (PSNR) or bit-rate performance, or sometimes both. This paper also proposes a modified algorithm based on a test-zone (TZ) search algorithm, a widely used fast-search algorithm with good rate-distortion performance, suitable for hardware implementation for encoding ultra-high-definition videos in real time. The results show that worst-case PSNR degradation and bit-rate increases compared with the TZ search in the HEVC reference software implementation are negligible, and the hardware gate count is less than for many other designs in the literature. [ABSTRACT FROM AUTHOR]

We investigate video classification via a two-stream convolutional neural network (CNN) design that directly ingests information extracted from compressed video bitstreams. Our approach begins with the observation that all modern video codecs divide the input frames into macroblocks (MBs). We demonstrate that selective access to MB motion vector (MV) information within compressed video bitstreams can also provide for selective, motion-adaptive, MB pixel decoding (a.k.a., MB texture decoding). This in turn allows for the derivation of spatio-temporal video activity regions at extremely high speed in comparison to conventional full-frame decoding followed by optical flow estimation. In order to evaluate the accuracy of a video classification framework based on such activity data, we independently train two CNN architectures on MB texture and MV correspondences and then fuse their scores to derive the final classification of each test video. Evaluation on two standard data sets shows that the proposed approach is competitive with the best two-stream video classification approaches found in the literature. At the same time: 1) a CPU-based realization of our MV extraction is over 977 times faster than GPU-based optical flow methods; 2) selective decoding is up to 12 times faster than full-frame decoding; and 3) our proposed spatial and temporal CNNs perform inference at 5 to 49 times lower cloud computing cost than the fastest methods from the literature. [ABSTRACT FROM AUTHOR]

A video compression framework based on spatio-temporal resolution adaptation (ViSTRA) is proposed, which dynamically resamples the input video spatially and temporally during encoding, based on a quantisation-resolution decision, and reconstructs the full resolution video at the decoder. Temporal upsampling is performed using frame repetition, whereas a convolutional neural network super-resolution model is employed for spatial resolution upsampling. ViSTRA has been integrated into the high efficiency video coding reference software (HM 16.14). Experimental results verified via an international challenge show significant improvements, with BD-rate gains of 15% based on PSNR and an average MOS difference of 0.5 based on subjective visual quality tests. [ABSTRACT FROM AUTHOR]

Recently, multiple ($k, n$) region-incrementing VCSs (RIVCSs) have been proposed that can gradually reconstruct secrets in a single image. In ($k, n$)-RIVCS, the secret image is subdivided into multiple regions in such a way that any $t$ shadow images, where $k \le t \le n$ , can be used to reveal the secret in the ($t-k+1$)th region. This region-incrementing property provides progressive decoding. However, the regions in previous RIVCSs are disjointed: no two regions have overlapping areas. In this paper, we discuss a ($k, n$) region-in-region progressive VCS (RPVCS). Our ($k, n$)-RPVCS has ($n-k+1$) secrecy-level regions, and each region can be located within any previous region. This new type of region allocation not only provides more areas in which to hide the secret compared with the non-overlapping regions in RIVCS but also provides different presentation methods in progressive decoding. [ABSTRACT FROM AUTHOR]

To save manufacturing cost, most color digital video cameras employ a single-sensor technology with a red–green–blue (RGB) color filter array (CFA) to capture real-world scenes. Due to only one primary color measured at each pixel location, the captured videos are usually referred to as the mosaic videos. For the purposes of economical storage and transmission, it is very important to achieve a good tradeoff between the quality and bitrate when compressing mosaic videos with different RGB-CFA structures. In this paper, based on mathematical optimization technique, a novel chroma subsampling strategy is presented for compressing mosaic videos with arbitrary RGB-CFA structures in H.264/AVC and High Efficiency Video Coding (HEVC). For each $2 \times 2$ YUV block to be subsampled with 4:2:0 format, the proposed strategy determines the proper sampled $U$ and $V$ components by minimizing, prior to compression, the quality distortion between the original colocated mosaic block and the mosaic block conversed from the current subsampled YUV block. Through the mathematical optimization formulated in the proposed strategy, the significance of the sampled $U$ and $V$ components for reconstructing R, G, and B pixels can be simultaneously taken into consideration. The experimental results demonstrate that the proposed chroma subsampling strategy has the best quality and bitrate tradeoff at a similar execution time requirement for compressing mosaic videos with arbitrary RGB-CFA structures in H.264/AVC and HEVC compared with the state-of-the-art ones by Chen et al. and Yang et al. as well as the three commonly used ones. [ABSTRACT FROM AUTHOR]

This paper provides an overview of Scalable High efficiency Video Coding (SHVC), the scalable extensions of the High Efficiency Video Coding (HEVC) standard, published in the second version of HEVC. In addition to the temporal scalability already provided by the first version of HEVC, SHVC further provides spatial, signal-to-noise ratio, bit depth, and color gamut scalability functionalities, as well as combinations of any of these. The SHVC architecture design enables SHVC implementations to be built using multiple repurposed single-layer HEVC codec cores, with the addition of interlayer reference picture processing modules. The general multilayer high-level syntax design common to all multilayer HEVC extensions, including SHVC, MV-HEVC, and 3D HEVC, is described. The interlayer reference picture processing modules, including texture and motion resampling and color mapping, are also described. Performance comparisons are provided for SHVC versus simulcast HEVC and versus the scalable video coding extension to H.264/advanced video coding. [ABSTRACT FROM PUBLISHER]

Saliency detection is widely used to extract regions of interest in images for various image processing applications. Recently, many saliency detection models have been proposed for video in uncompressed (pixel) domain. However, video over Internet is always stored in compressed domains, such as MPEG2, H.264, and MPEG4 Visual. In this paper, we propose a novel video saliency detection model based on feature contrast in compressed domain. Four types of features including luminance, color, texture, and motion are extracted from the discrete cosine transform coefficients and motion vectors in video bitstream. The static saliency map of unpredicted frames (I frames) is calculated on the basis of luminance, color, and texture features, while the motion saliency map of predicted frames (P and B frames) is computed by motion feature. A new fusion method is designed to combine the static saliency and motion saliency maps to get the final saliency map for each video frame. Due to the directly derived features in compressed domain, the proposed model can predict the salient regions efficiently for video frames. Experimental results on a public database show superior performance of the proposed video saliency detection model in compressed domain. [ABSTRACT FROM AUTHOR]

This paper presents a method to reduce the computation and memory access for variable block size motion estimation (ME) using pixel truncation. Previous work has focused on implementing pixel truncation using a fixed-block- size (16 x 16 pixels) ME. However, pixel truncation fails to give satisfactory results for smaller block partitions. In this paper, we analyze the effect of truncating pixels for smaller block partitions and propose a method to improve the frame prediction. Our method is able to reduce the total computation and memory access compared to conventional full-search method without significantly degrading picture quality. With unique data arrangement, the proposed architectures are able to save up to 53% energy compared to the conventional full-search architecture. This makes such architectures attractive for H.264 application in future mobile devices. [ABSTRACT FROM AUTHOR]

Subpixel motion estimation plays an important role in compression efficiency within modern video codecs such as MPEG2, MPEG4, and H.264. Subpixel motion estimation is implemented within these standards using interpolated values at 1/2 or 1/4 pixel accuracy. Such interpolation gives a good reduction in residual energy for each predicted macroblock and, therefore, improves compression. However, this leads to a significant increase in computational complexity at the encoder. This is especially true for H.264 where the cost of an exhaustive set of macroblock segmentations need to be estimated in order to obtain an optimal mode for prediction. This paper presents a novel interpolation-free scheme for subpixel motion estimation using the result of the full pixel sum of absolute difference distribution of each motion compensated block applied to an H.264 encoder. This system produces reduced complexity motion estimation with a controllable tradeoff between compression performance and encoder speed. These methods facilitate the generation of a real time software H.264 encoder. [ABSTRACT FROM AUTHOR]