We present the first neural video compression method based on generative adversarial networks (GANs). Our approach significantly outperforms previous neural and non-neural video compression methods in a user study, setting a new state-of-the-art in visual quality for neural methods. We show that the GAN loss is crucial to obtain this high visual quality. Two components make the GAN loss effective: we i) synthesize detail by conditioning the generator on a latent extracted from the warped previous reconstruction to then ii) propagate this detail with high-quality flow. We find that user studies are required to compare methods, i.e., none of our quantitative metrics were able to predict all studies. We present the network design choices in detail, and ablate them with user studies.
Quality assessment plays a key role in creating and comparing video compression algorithms. Despite the development of a large number of new methods for assessing quality, generally accepted and well-known codecs comparisons mainly use the classical methods like PSNR, SSIM and new method VMAF. These methods can be calculated following different rules: they can use different frame-by-frame averaging techniques or different summation of color components. In this paper, a fundamental comparison of various versions of generally accepted metrics is carried out to find the most relevant and recommended versions of video quality metrics to be used in codecs comparisons. For comparison, we used a set of videos encoded with video codecs of different standards, and visual quality scores collected for the resulting set of streams since 2018 until 2021
COVID-19 leads to the high demand for remote interactive systems ever seen. One of the key elements of these systems is video streaming, which requires a very high network bandwidth due to its specific real-time demand, especially with high-resolution video. Existing video compression methods are struggling in the trade-off between video quality and the speed requirement. Addressed that the background information rarely changes in most remote meeting cases, we introduce a Region-Of-Interests (ROI) based video compression framework (named RCLC) that leverages the cutting-edge learning-based and conventional technologies. In RCLC, each coming frame is marked as a background-updating (BU) or ROI-updating (RU) frame. By applying the conventional video codec, the BU frame is compressed with low-quality and high-compression, while the ROI from RU-frame is compressed with high-quality and low-compression. The learning-based methods are applied to detect the ROI, blend background-ROI, and enhance video quality. The experimental results show that our RCLC can reduce up to 32.55\% BD-rate for the ROI region compared to H.265 video codec under a similar compression time with 1080p resolution.
Video represents the majority of internet traffic today, driving a continual race between the generation of higher quality content, transmission of larger file sizes, and the development of network infrastructure. In addition, the recent COVID-19 pandemic fueled a surge in the use of video conferencing tools. Since videos take up considerable bandwidth (~100 Kbps to a few Mbps), improved video compression can have a substantial impact on network performance for live and pre-recorded content, providing broader access to multimedia content worldwide. We present a novel video compression pipeline, called Txt2Vid, which dramatically reduces data transmission rates by compressing webcam videos ("talking-head videos") to a text transcript. The text is transmitted and decoded into a realistic reconstruction of the original video using recent advances in deep learning based voice cloning and lip syncing models. Our generative pipeline achieves two to three orders of magnitude reduction in the bitrate as compared to the standard audio-video codecs (encoders-decoders), while maintaining equivalent Quality-of-Experience based on a subjective evaluation by users (n = 242) in an online study. The Txt2Vid framework opens up the potential for creating novel applications such as enabling audio-video communication during poor internet connectivity, or in remote terrains with limited bandwidth. The code for this work is available at https://github.com/tpulkit/txt2vid.git.
The paper presents a new approach to multiview video coding using Screen Content Coding. It is assumed that for a time instant the frames corresponding to all views are packed into a single frame, i.e. the frame-compatible approach to multiview coding is applied. For such coding scenario, the paper demonstrates that Screen Content Coding can be efficiently used for multiview video coding. Two approaches are considered: the first using standard HEVC Screen Content Coding, and the second using Advanced Screen Content Coding. The latter is the original proposal of the authors that exploits quarter-pel motion vectors and other nonstandard extensions of HEVC Screen Content Coding. The experimental results demonstrate that multiview video coding even using standard HEVC Screen Content Coding is much more efficient than simulcast HEVC coding. The proposed Advanced Screen Content Coding provides virtually the same coding efficiency as MV-HEVC, which is the state-of-the-art multiview video compression technique. The authors suggest that Advanced Screen Content Coding can be efficiently used within the new Versatile Video Coding (VVC) technology. Nevertheless a reference multiview extension of VVC does not exist yet, therefore, for VVC-based coding, the experimental comparisons are left for future work.
Previous blind or No Reference (NR) video quality assessment (VQA) models largely rely on features drawn from natural scene statistics (NSS), but under the assumption that the image statistics are stationary in the spatial domain. Several of these models are quite successful on standard pictures. However, in Virtual Reality (VR) applications, foveated video compression is regaining attention, and the concept of space-variant quality assessment is of interest, given the availability of increasingly high spatial and temporal resolution contents and practical ways of measuring gaze direction. Distortions from foveated video compression increase with increased eccentricity, implying that the natural scene statistics are space-variant. Towards advancing the development of foveated compression / streaming algorithms, we have devised a no-reference (NR) foveated video quality assessment model, called FOVQA, which is based on new models of space-variant natural scene statistics (NSS) and natural video statistics (NVS). Specifically, we deploy a space-variant generalized Gaussian distribution (SV-GGD) model and a space-variant asynchronous generalized Gaussian distribution (SV-AGGD) model of mean subtracted contrast normalized (MSCN) coefficients and products of neighboring MSCN coefficients, respectively. We devise a foveated video quality predictor that extracts radial basis features, and other features that capture perceptually annoying rapid quality fall-offs. We find that FOVQA achieves state-of-the-art (SOTA) performance on the new 2D LIVE-FBT-FCVR database, as compared with other leading FIQA / VQA models. we have made our implementation of FOVQA available at: http://live.ece.utexas.edu/research/Quality/FOVQA.zip.
Visual analytics have played an increasingly critical role in the Internet of Things, where massive visual signals have to be compressed and fed into machines. But facing such big data and constrained bandwidth capacity, existing image/video compression methods lead to very low-quality representations, while existing feature compression techniques fail to support diversified visual analytics applications/tasks with low-bit-rate representations. In this paper, we raise and study the novel problem of supporting multiple machine vision analytics tasks with the compressed visual representation, namely, the information compression problem in analytics taxonomy. By utilizing the intrinsic transferability among different tasks, our framework successfully constructs compact and expressive representations at low bit-rates to support a diversified set of machine vision tasks, including both high-level semantic-related tasks and mid-level geometry analytic tasks. In order to impose compactness in the representations, we propose a codebook-based hyperprior, which helps map the representation into a low-dimensional manifold. As it well fits the signal structure of the deep visual feature, it facilitates more accurate entropy estimation, and results in higher compression efficiency. With the proposed framework and the codebook-based hyperprior, we further investigate the relationship of different task features owning different levels of abstraction granularity. Experimental results demonstrate that with the proposed scheme, a set of diversified tasks can be supported at a significantly lower bit-rate, compared with existing compression schemes.
The versatility of recent machine learning approaches makes them ideal for improvement of next generation video compression solutions. Unfortunately, these approaches typically bring significant increases in computational complexity and are difficult to interpret into explainable models, affecting their potential for implementation within practical video coding applications. This paper introduces a novel explainable neural network-based inter-prediction scheme, to improve the interpolation of reference samples needed for fractional precision motion compensation. The approach requires a single neural network to be trained from which a full quarter-pixel interpolation filter set is derived, as the network is easily interpretable due to its linear structure. A novel training framework enables each network branch to resemble a specific fractional shift. This practical solution makes it very efficient to use alongside conventional video coding schemes. When implemented in the context of the state-of-the-art Versatile Video Coding (VVC) test model, 0.77%, 1.27% and 2.25% BD-rate savings can be achieved on average for lower resolution sequences under the random access, low-delay B and low-delay P configurations, respectively, while the complexity of the learned interpolation schemes is significantly reduced compared to the interpolation with full CNNs.
This paper describes an adaptive Lagrange multiplier determination method for rate-quality optimisation in video compression. Inspired by the experimental results of a Lagrange multiplier selection test, the presented approach adaptively estimates the optimum Lagrange multiplier for different video content, based on distortion statistics of recently encoded frames. The proposed algorithm has been fully integrated into both the H.264 and HEVC reference codecs, and is used in rate-distortion optimisation for encoding B frames. The results show promising (up to 11% on the sequences tested) overall bitrate savings, for a minimal increase in complexity, on various types of test content based on Bjontegaard delta measurements.
This paper describes a quality assessment model for perceptual video compression applications (PVM), which stimulates visual masking and distortion-artefact perception using an adaptive combination of noticeable distortions and blurring artefacts. The method shows significant improvement over existing quality metrics based on the VQEG database, and provides compatibility with in-loop rate-quality optimisation for next generation video codecs due to its latency and complexity attributes. Performance comparison are validated against a range of different distortion types.
Spatial resolution adaptation is a technique which has often been employed in video compression to enhance coding efficiency. This approach encodes a lower resolution version of the input video and reconstructs the original resolution during decoding. Instead of using conventional up-sampling filters, recent work has employed advanced super-resolution methods based on convolutional neural networks (CNNs) to further improve reconstruction quality. These approaches are usually trained to minimise pixel-based losses such as Mean-Squared Error (MSE), despite the fact that this type of loss metric does not correlate well with subjective opinions. In this paper, a perceptually-inspired super-resolution approach (M-SRGAN) is proposed for spatial up-sampling of compressed video using a modified CNN model, which has been trained using a generative adversarial network (GAN) on compressed content with perceptual loss functions. The proposed method was integrated with HEVC HM 16.20, and has been evaluated on the JVET Common Test Conditions (UHD test sequences) using the Random Access configuration. The results show evident perceptual quality improvement over the original HM 16.20, with an average bitrate saving of 35.6% (Bj{\o}ntegaard Delta measurement) based on a perceptual quality metric, VMAF.
Virtual Reality is regaining attention due to recent advancements in hardware technology. Immersive images / videos are becoming widely adopted to carry omnidirectional visual information. However, due to the requirements for higher spatial and temporal resolution of real video data, immersive videos require significantly larger bandwidth consumption. To reduce stresses on bandwidth, foveated video compression is regaining popularity, whereby the space-variant spatial resolution of the retina is exploited. Towards advancing the progress of foveated video compression, we propose a full reference (FR) foveated image quality assessment algorithm, which we call foveated entropic differencing (FED), which employs the natural scene statistics of bandpass responses by applying differences of local entropies weighted by a foveation-based error sensitivity function. We evaluate the proposed algorithm by measuring the correlations of the predictions that FED makes against human judgements on the newly created 2D and 3D LIVE-FBT-FCVR databases for Virtual Reality (VR). The performance of the proposed algorithm yields state-of-the-art as compared with other existing full reference algorithms. Software for FED has been made available at: http://live.ece.utexas.edu/research/Quality/FED.zip
Structural similarity (SSIM)-based distortion Dssim is more consistent with human perception than the traditional mean squared error Dmse. To achieve better video encoding quality, many studies on optimal bit allocation (OBA) used Dssim as the distortion metric. However, the MSE-based rate distortion optimization (RDO) was still used in these studies. The inconsistency between the optimization goals of OBA and RDO results in a non-optimal SSIM-based encoding performance. To solve this problem, we propose an accurate coding tree unit level Dssim-Dmse model, which enables performing the SSIM-based RDO with simpler R-Dmse cost scaled by the SSIM-based Lagrangian parameter λssim. Moreover, based on this model, the R-Dssim model can be accurately estimated based on the joint relationship of R-Dssim-λssim. With the accurate R-Dssim model, the SSIM-based OBA problem is then solved. Accordingly, the SSIM-based OBA and SSIM-based RDO are unified together in our scheme, called SOSR. Compared with the HEVC reference encoder HM16.20, SOSR saves 5%, 11%, and 17% bitrate under the same SSIM in the commonly used all-intra, hierarchical and non-hierarchical low-delay-B configurations, which is superior to existing state-of-the-art SSIM-based OBA schemes.
Many deep learning based video compression artifact removal algorithms have been proposed to recover high-quality videos from low-quality compressed videos. Recently, methods were proposed to mine spatiotemporal information via utilizing multiple neighboring frames as reference frames. However, these post-processing methods take advantage of adjacent frames directly, but neglect the information of the video itself, which can be exploited. In this paper, we propose an effective reference frame proposal strategy to boost the performance of the existing multi-frame approaches. Besides, we introduce a loss based on fast Fourier transformation~(FFT) to further improve the effectiveness of restoration. Experimental results show that our method achieves better fidelity and perceptual performance on MFQE 2.0 dataset than the state-of-the-art methods. And our method won Track 1 and Track 2, and was ranked the 2nd in Track 3 of NTIRE 2021 Quality enhancement of heavily compressed videos Challenge.
Learned frame prediction is a current problem of interest in computer vision and video compression. Although several deep network architectures have been proposed for learned frame prediction, to the best of our knowledge, there is no work based on using deformable convolutions for frame prediction. To this effect, we propose a deformable frame prediction network (DFPN) for task oriented implicit motion modeling and next frame prediction. Experimental results demonstrate that the proposed DFPN model achieves state of the art results in next frame prediction. Our models and results are available at https://github.com/makinyilmaz/DFPN.
Machine learning techniques for more efficient video compression and video enhancement have been developed thanks to breakthroughs in deep learning. The new techniques, considered as an advanced form of Artificial Intelligence (AI), bring previously unforeseen capabilities. However, they typically come in the form of resource-hungry black-boxes (overly complex with little transparency regarding the inner workings). Their application can therefore be unpredictable and generally unreliable for large-scale use (e.g. in live broadcast). The aim of this work is to understand and optimise learned models in video processing applications so systems that incorporate them can be used in a more trustworthy manner. In this context, the presented work introduces principles for simplification of learned models targeting improved transparency in implementing machine learning for video production and distribution applications. These principles are demonstrated on video compression examples, showing how bitrate savings and reduced complexity can be achieved by simplifying relevant deep learning models.
The layers of convolutional neural networks (CNNs) can be used to alter the resolution of their inputs, but the scaling factors are limited to integer values. However, in many image and video processing applications, the ability to resize by a fractional factor would be advantageous. One example is conversion between resolutions standardized for video compression, such as from 1080p to 720p. To solve this problem, we propose an alternative building block, formulated as a conventional convolutional layer followed by a differentiable resizer. More concretely, the convolutional layer preserves the resolution of the input, while the resizing operation is fully handled by the resizer. In this way, any CNN architecture can be adapted for non-integer resizing. As an application, we replace the resizing convolutional layer of a modern deep downsampling model by the proposed building block, and apply it to an adaptive bitrate video streaming scenario. Our experimental results show that an improvement in coding efficiency over the conventional Lanczos algorithm is attained, in terms of PSNR, SSIM, and VMAF on test videos.
Learning based video compression attracts increasing attention in the past few years. The previous hybrid coding approaches rely on pixel space operations to reduce spatial and temporal redundancy, which may suffer from inaccurate motion estimation or less effective motion compensation. In this work, we propose a feature-space video coding network (FVC) by performing all major operations (i.e., motion estimation, motion compression, motion compensation and residual compression) in the feature space. Specifically, in the proposed deformable compensation module, we first apply motion estimation in the feature space to produce motion information (i.e., the offset maps), which will be compressed by using the auto-encoder style network. Then we perform motion compensation by using deformable convolution and generate the predicted feature. After that, we compress the residual feature between the feature from the current frame and the predicted feature from our deformable compensation module. For better frame reconstruction, the reference features from multiple previous reconstructed frames are also fused by using the non-local attention mechanism in the multi-frame feature fusion module. Comprehensive experimental results demonstrate that the proposed framework achieves the state-of-the-art performance on four benchmark datasets including HEVC, UVG, VTL and MCL-JCV.
Rate-distortion optimization (RDO) of codecs, where distortion is quantified by the mean-square error, has been a standard practice in image/video compression over the years. RDO serves well for optimization of codec performance for evaluation of the results in terms of PSNR. However, it is well known that the PSNR does not correlate well with perceptual evaluation of images; hence, RDO is not well suited for perceptual optimization of codecs. Recently, rate-distortion-perception trade-off has been formalized by taking the Kullback-Leibner (KL) divergence between the distributions of the original and reconstructed images as a perception measure. Learned image compression methods that simultaneously optimize rate, mean-square loss, VGG loss, and an adversarial loss were proposed. Yet, there exists no easy approach to fix the rate, distortion or perception at a desired level in a practical learned image compression solution to perform an analysis of the trade-off between rate, distortion and perception measures. In this paper, we propose a practical approach to fix the rate to carry out perception-distortion analysis at a fixed rate in order to perform perceptual evaluation of image compression results in a principled manner. Experimental results provide several insights for practical rate-distortion-perception analysis in learned image compression.
While learned video codecs have demonstrated great promise, they have yet to achieve sufficient efficiency for practical deployment. In this work, we propose several novel ideas for learned video compression which allow for improved performance for the low-latency mode (I- and P-frames only) along with a considerable increase in computational efficiency. In this setting, for natural videos our approach compares favorably across the entire R-D curve under metrics PSNR, MS-SSIM and VMAF against all mainstream video standards (H.264, H.265, AV1) and all ML codecs. At the same time, our approach runs at least 5x faster and has fewer parameters than all ML codecs which report these figures. Our contributions include a flexible-rate framework allowing a single model to cover a large and dense range of bitrates, at a negligible increase in computation and parameter count; an efficient backbone optimized for ML-based codecs; and a novel in-loop flow prediction scheme which leverages prior information towards more efficient compression. We benchmark our method, which we call ELF-VC (Efficient, Learned and Flexible Video Coding) on popular video test sets UVG and MCL-JCV under metrics PSNR, MS-SSIM and VMAF. For example, on UVG under PSNR, it reduces the BD-rate by 44% against H.264, 26% against H.265, 15% against AV1, and 35% against the current best ML codec. At the same time, on an NVIDIA Titan V GPU our approach encodes/decodes VGA at 49/91 FPS, HD 720 at 19/35 FPS, and HD 1080 at 10/18 FPS.