While Large Language Models (LLMs) are the dominant models for generative tasks in language, they do not perform as well as diffusion models on image and video generation. To effectively use LLMs for visual generation, one crucial component is the visual tokenizer that maps pixel-space inputs to discrete tokens appropriate for LLM learning. In this paper, we introduce MAGVIT-v2, a video tokenizer designed to generate concise and expressive tokens for both videos and images using a common token vocabulary. Equipped with this new tokenizer, we show that LLMs outperform diffusion models on standard image and video generation benchmarks including ImageNet and Kinetics. In addition, we demonstrate that our tokenizer surpasses the previously top-performing video tokenizer on two more tasks: (1) video compression comparable to the next-generation video codec (VCC) according to human evaluations, and (2) learning effective representations for action recognition tasks.
With the growing demand for video applications, many advanced learned video compression methods have been developed, outperforming traditional methods in terms of objective quality metrics such as PSNR. Existing methods primarily focus on objective quality but tend to overlook perceptual quality. Directly incorporating perceptual loss into a learned video compression framework is nontrivial and raises several perceptual quality issues that need to be addressed. In this paper, we investigated these issues in learned video compression and propose a novel High Visual-Fidelity Learned Video Compression framework (HVFVC). Specifically, we design a novel confidence-based feature reconstruction method to address the issue of poor reconstruction in newly-emerged regions, which significantly improves the visual quality of the reconstruction. Furthermore, we present a periodic compensation loss to mitigate the checkerboard artifacts related to deconvolution operation and optimization. Extensive experiments have shown that the proposed HVFVC achieves excellent perceptual quality, outperforming the latest VVC standard with only 50% required bitrate.
Neural video codecs have recently become competitive with standard codecs such as HEVC in the low-delay setting. However, most neural codecs are large floating-point networks that use pixel-dense warping operations for temporal modeling, making them too computationally expensive for deployment on mobile devices. Recent work has demonstrated that running a neural decoder in real time on mobile is feasible, but shows this only for 720p RGB video. This work presents the first neural video codec that decodes 1080p YUV420 video in real time on a mobile device. Our codec relies on two major contributions. First, we design an efficient codec that uses a block-based motion compensation algorithm available on the warping core of the mobile accelerator, and we show how to quantize this model to integer precision. Second, we implement a fast decoder pipeline that concurrently runs neural network components on the neural signal processor, parallel entropy coding on the mobile GPU, and warping on the warping core. Our codec outperforms the previous on-device codec by a large margin with up to 48% BD-rate savings, while reducing the MAC count on the receiver side by $10 \times$. We perform a careful ablation to demonstrate the effect of the introduced motion compensation scheme, and ablate the effect of model quantization.
Finding groups of similar image blocks within an ample search area is often necessary in different applications, such as video compression, image clustering, vector quantization, and nonlocal noise reduction. A block-matching algorithm that uses a dissimilarity measure can be applied in such scenarios. In this work, a measure that utilizes the quantum Fourier transform or the Swap test based on the Euclidean distance is proposed. Experiments on small cases with ideal and noisy simulations are implemented. In the case of the Swap test, the IBM and IonQ quantum devices have been used, demonstrating potential for future near-term applications.
Learned B-frame video compression aims to adopt bi-directional motion estimation and motion compensation (MEMC) coding for middle frame reconstruction. However, previous learned approaches often directly extend neural P-frame codecs to B-frame relying on bi-directional optical-flow estimation or video frame interpolation. They suffer from inaccurate quantized motions and inefficient motion compensation. To address these issues, we propose a simple yet effective structure called Interpolation-driven B-frame Video Compression (IBVC). Our approach only involves two major operations: video frame interpolation and artifact reduction compression. IBVC introduces a bit-rate free MEMC based on interpolation, which avoids optical-flow quantization and additional compression distortions. Later, to reduce duplicate bit-rate consumption and focus on unaligned artifacts, a residual guided masking encoder is deployed to adaptively select the meaningful contexts with interpolated multi-scale dependencies. In addition, a conditional spatio-temporal decoder is proposed to eliminate location errors and artifacts instead of using MEMC coding in other methods. The experimental results on B-frame coding demonstrate that IBVC has significant improvements compared to the relevant state-of-the-art methods. Meanwhile, our approach can save bit rates compared with the random access (RA) configuration of H.266 (VTM). The code will be available at https://github.com/ruhig6/IBVC.
Learned video compression (LVC) has witnessed remarkable advancements in recent years. Similar as the traditional video coding, LVC inherits motion estimation/compensation, residual coding and other modules, all of which are implemented with neural networks (NNs). However, within the framework of NNs and its training mechanism using gradient backpropagation, most existing works often struggle to consistently generate stable motion information, which is in the form of geometric features, from the input color features. Moreover, the modules such as the inter-prediction and residual coding are independent from each other, making it inefficient to fully reduce the spatial-temporal redundancy. To address the above problems, in this paper, we propose a novel Spatial-Temporal Transformer based Video Compression (STT-VC) framework. It contains a Relaxed Deformable Transformer (RDT) with Uformer based offsets estimation for motion estimation and compensation, a Multi-Granularity Prediction (MGP) module based on multi-reference frames for prediction refinement, and a Spatial Feature Distribution prior based Transformer (SFD-T) for efficient temporal-spatial joint residual compression. Specifically, RDT is developed to stably estimate the motion information between frames by thoroughly investigating the relationship between the similarity based geometric motion feature extraction and self-attention. MGP is designed to fuse the multi-reference frame information by effectively exploring the coarse-grained prediction feature generated with the coded motion information. SFD-T is to compress the residual information by jointly exploring the spatial feature distributions in both residual and temporal prediction to further reduce the spatial-temporal redundancy. Experimental results demonstrate that our method achieves the best result with 13.5% BD-Rate saving over VTM.
NASA's Solar Dynamics Observatory (SDO) mission collects large data volumes of the Sun's daily activity. Data compression is crucial for space missions to reduce data storage and video bandwidth requirements by eliminating redundancies in the data. In this paper, we present a novel neural Transformer-based video compression approach specifically designed for the SDO images. Our primary objective is to efficiently exploit the temporal and spatial redundancies inherent in solar images to obtain a high compression ratio. Our proposed architecture benefits from a novel Transformer block called Fused Local-aware Window (FLaWin), which incorporates window-based self-attention modules and an efficient fused local-aware feed-forward (FLaFF) network. This architectural design allows us to simultaneously capture short-range and long-range information while facilitating the extraction of rich and diverse contextual representations. Moreover, this design choice results in reduced computational complexity. Experimental results demonstrate the significant contribution of the FLaWin Transformer block to the compression performance, outperforming conventional hand-engineered video codecs such as H.264 and H.265 in terms of rate-distortion trade-off.
Commonly used datasets for evaluating video codecs are all very high quality and not representative of video typically used in video conferencing scenarios. We present the Video Conferencing Dataset (VCD) for evaluating video codecs for real-time communication, the first such dataset focused on video conferencing. VCD includes a wide variety of camera qualities and spatial and temporal information. It includes both desktop and mobile scenarios and two types of video background processing. We report the compression efficiency of H.264, H.265, H.266, and AV1 in low-delay settings on VCD and compare it with the non-video conferencing datasets UVC, MLC-JVC, and HEVC. The results show the source quality and the scenarios have a significant effect on the compression efficiency of all the codecs. VCD enables the evaluation and tuning of codecs for this important scenario. The VCD is publicly available as an open-source dataset at https://github.com/microsoft/VCD.
The past decade has witnessed the huge success of deep learning in well-known artificial intelligence applications such as face recognition, autonomous driving, and large language model like ChatGPT. Recently, the application of deep learning has been extended to a much wider range, with neural network-based video coding being one of them. Neural network-based video coding can be performed at two different levels: embedding neural network-based (NN-based) coding tools into a classical video compression framework or building the entire compression framework upon neural networks. This paper elaborates some of the recent exploration efforts of JVET (Joint Video Experts Team of ITU-T SG 16 WP 3 and ISO/IEC JTC 1/SC29) in the name of neural network-based video coding (NNVC), falling in the former category. Specifically, this paper discusses two major NN-based video coding technologies, i.e. neural network-based intra prediction and neural network-based in-loop filtering, which have been investigated for several meeting cycles in JVET and finally adopted into the reference software of NNVC. Extensive experiments on top of the NNVC have been conducted to evaluate the effectiveness of the proposed techniques. Compared with VTM-11.0_nnvc, the proposed NN-based coding tools in NNVC-4.0 could achieve {11.94%, 21.86%, 22.59%}, {9.18%, 19.76%, 20.92%}, and {10.63%, 21.56%, 23.02%} BD-rate reductions on average for {Y, Cb, Cr} under random-access, low-delay, and all-intra configurations respectively.
Standardized lossy video coding is at the core of almost all real-world video processing pipelines. Rate control is used to enable standard codecs to adapt to different network bandwidth conditions or storage constraints. However, standard video codecs (e.g., H.264) and their rate control modules aim to minimize video distortion w.r.t. human quality assessment. We demonstrate empirically that standard-coded videos vastly deteriorate the performance of deep vision models. To overcome the deterioration of vision performance, this paper presents the first end-to-end learnable deep video codec control that considers both bandwidth constraints and downstream deep vision performance, while adhering to existing standardization. We demonstrate that our approach better preserves downstream deep vision performance than traditional standard video coding.
Neural video compression (NVC) is a rapidly evolving video coding research area, with some models achieving superior coding efficiency compared to the latest video coding standard Versatile Video Coding (VVC). In conventional video coding standards, the hierarchical B-frame coding, which utilizes a bidirectional prediction structure for higher compression, had been well-studied and exploited. In NVC, however, limited research has investigated the hierarchical B scheme. In this paper, we propose an NVC model exploiting hierarchical B-frame coding with temporal layer-adaptive optimization. We first extend an existing unidirectional NVC model to a bidirectional model, which achieves -21.13% BD-rate gain over the unidirectional baseline model. However, this model faces challenges when applied to sequences with complex or large motions, leading to performance degradation. To address this, we introduce temporal layer-adaptive optimization, incorporating methods such as temporal layer-adaptive quality scaling (TAQS) and temporal layer-adaptive latent scaling (TALS). The final model with the proposed methods achieves an impressive BD-rate gain of -39.86% against the baseline. It also resolves the challenges in sequences with large or complex motions with up to -49.13% more BD-rate gains than the simple bidirectional extension. This improvement is attributed to the allocation of more bits to lower temporal layers, thereby enhancing overall reconstruction quality with smaller bits. Since our method has little dependency on a specific NVC model architecture, it can serve as a general tool for extending unidirectional NVC models to the ones with hierarchical B-frame coding.
Multimedia compression allows us to watch videos, see pictures and hear sounds within a limited bandwidth, which helps the flourish of the internet. During the past decades, multimedia compression has achieved great success using hand-craft features and systems. With the development of artificial intelligence and video compression, there emerges a lot of research work related to using the neural network on the video compression task to get rid of the complicated system. Not only producing the advanced algorithms, but researchers also spread the compression to different content, such as User Generated Content(UGC). With the rapid development of mobile devices, screen content videos become an important part of multimedia data. In contrast, we find community lacks a large-scale dataset for screen content video compression, which impedes the fast development of the corresponding learning-based algorithms. In order to fulfill this blank and accelerate the research of this special type of videos, we propose the Large-scale Screen Content Dataset(LSCD), which contains 714 source sequences. Meanwhile, we provide the analysis of the proposed dataset to show some features of screen content videos, which will help researchers have a better understanding of how to explore new algorithms. Besides collecting and post-processing the data to organize the dataset, we also provide a benchmark containing the performance of both traditional codec and learning-based methods.
The proliferation of technologies, such as extended reality (XR), has increased the demand for high-quality three-dimensional (3D) graphical representations. Industrial 3D applications encompass computer-aided design (CAD), finite element analysis (FEA), scanning, and robotics. However, current methods employed for industrial 3D representations suffer from high implementation costs and reliance on manual human input for accurate 3D modeling. To address these challenges, neural radiance fields (NeRFs) have emerged as a promising approach for learning 3D scene representations based on provided training 2D images. Despite a growing interest in NeRFs, their potential applications in various industrial subdomains are still unexplored. In this paper, we deliver a comprehensive examination of NeRF industrial applications while also providing direction for future research endeavors. We also present a series of proof-of-concept experiments that demonstrate the potential of NeRFs in the industrial domain. These experiments include NeRF-based video compression techniques and using NeRFs for 3D motion estimation in the context of collision avoidance. In the video compression experiment, our results show compression savings up to 48\% and 74\% for resolutions of 1920x1080 and 300x168, respectively. The motion estimation experiment used a 3D animation of a robotic arm to train Dynamic-NeRF (D-NeRF) and achieved an average peak signal-to-noise ratio (PSNR) of disparity map with the value of 23 dB and an structural similarity index measure (SSIM) 0.97.
Deepfake detection methods have shown promising results in recognizing forgeries within a given dataset, where training and testing take place on the in-distribution dataset. However, their performance deteriorates significantly when presented with unseen samples. As a result, a reliable deepfake detection system must remain impartial to forgery types, appearance, and quality for guaranteed generalizable detection performance. Despite various attempts to enhance cross-dataset generalization, the problem remains challenging, particularly when testing against common post-processing perturbations, such as video compression or blur. Hence, this study introduces a deepfake detection framework, leveraging a self-supervised pre-training model that delivers exceptional generalization ability, withstanding common corruptions and enabling feature explainability. The framework comprises three key components: a feature extractor based on vision Transformer architecture that is pre-trained via self-supervised contrastive learning methodology, a graph convolution network coupled with a Transformer discriminator, and a graph Transformer relevancy map that provides a better understanding of manipulated regions and further explains the model's decision. To assess the effectiveness of the proposed framework, several challenging experiments are conducted, including in-data distribution performance, cross-dataset, cross-manipulation generalization, and robustness against common post-production perturbations. The results achieved demonstrate the remarkable effectiveness of the proposed deepfake detection framework, surpassing the current state-of-the-art approaches.
Conditional coding is a new video coding paradigm enabled by neural-network-based compression. It can be shown that conditional coding is in theory better than the traditional residual coding, which is widely used in video compression standards like HEVC or VVC. However, on closer inspection, it becomes clear that conditional coders can suffer from information bottlenecks in the prediction path, i.e., that due to the data processing inequality not all information from the prediction signal can be passed to the reconstructed signal, thereby impairing the coder performance. In this paper we propose the conditional residual coding concept, which we derive from information theoretical properties of the conditional coder. This coder significantly reduces the influence of bottlenecks, while maintaining the theoretical performance of the conditional coder. We provide a theoretical analysis of the coding paradigm and demonstrate the performance of the conditional residual coder in a practical example. We show that conditional residual coders alleviate the disadvantages of conditional coders while being able to maintain their advantages over residual coders. In the spectrum of residual and conditional coding, we can therefore consider them as ``the best from both worlds''.
Video analytics are often performed as cloud services in edge settings, mainly to offload computation, and also in situations where the results are not directly consumed at the video sensors. Sending high-quality video data from the edge devices can be expensive both in terms of bandwidth and power use. In order to build a streaming video analytics pipeline that makes efficient use of these resources, it is therefore imperative to reduce the size of the video stream. Traditional video compression algorithms are unaware of the semantics of the video, and can be both inefficient and harmful for the analytics performance. In this paper, we introduce LtC, a collaborative framework between the video source and the analytics server, that efficiently learns to reduce the video streams within an analytics pipeline. Specifically, LtC uses the full-fledged analytics algorithm at the server as a teacher to train a lightweight student neural network, which is then deployed at the video source. The student network is trained to comprehend the semantic significance of various regions within the videos, which is used to differentially preserve the crucial regions in high quality while the remaining regions undergo aggressive compression. Furthermore, LtC also incorporates a novel temporal filtering algorithm based on feature-differencing to omit transmitting frames that do not contribute new information. Overall, LtC is able to use 28-35% less bandwidth and has up to 45% shorter response delay compared to recently published state of the art streaming frameworks while achieving similar analytics performance.
Video coding has traditionally been developed to support services such as video streaming, videoconferencing, digital TV, and so on. The main intent was to enable human viewing of the encoded content. However, with the advances in deep neural networks (DNNs), encoded video is increasingly being used for automatic video analytics performed by machines. In applications such as automatic traffic monitoring, analytics such as vehicle detection, tracking and counting, would run continuously, while human viewing could be required occasionally to review potential incidents. To support such applications, a new paradigm for video coding is needed that will facilitate efficient representation and compression of video for both machine and human use in a scalable manner. In this manuscript, we introduce an end-to-end learnable video codec that supports a machine vision task in its base layer, while its enhancement layer, together with the base layer, supports input reconstruction for human viewing. The proposed system is constructed based on the concept of conditional coding to achieve better compression gains. Comprehensive experimental evaluations conducted on four standard video datasets demonstrate that our framework outperforms both state-of-the-art learned and conventional video codecs in its base layer, while maintaining comparable performance on the human vision task in its enhancement layer. Implementation of the proposed system is available at https://github.com/hadipardis/svc
Adaptive block partitioning is responsible for large gains in current image and video compression systems. This method is able to compress large stationary image areas with only a few symbols, while maintaining a high level of quality in more detailed areas. Current state-of-the-art neural-network-based image compression systems however use only one scale to transmit the latent space. In previous publications, we proposed RDONet, a scheme to transmit the latent space in multiple spatial resolutions. Following this principle, we extend a state-of-the-art compression network by a second hierarchical latent-space level to enable multi-scale processing. We extend the existing rate variability capabilities of RDONet by a gain unit. With that we are able to outperform an equivalent traditional autoencoder by 7% rate savings. Furthermore, we show that even though we add an additional latent space, the complexity only increases marginally and the decoding time can potentially even be decreased.
Video compression relies heavily on exploiting the temporal redundancy between video frames, which is usually achieved by estimating and using the motion information. The motion information is represented as optical flows in most of the existing deep video compression networks. Indeed, these networks often adopt pre-trained optical flow estimation networks for motion estimation. The optical flows, however, may be less suitable for video compression due to the following two factors. First, the optical flow estimation networks were trained to perform inter-frame prediction as accurately as possible, but the optical flows themselves may cost too many bits to encode. Second, the optical flow estimation networks were trained on synthetic data, and may not generalize well enough to real-world videos. We address the twofold limitations by enhancing the optical flows in two stages: offline and online. In the offline stage, we fine-tune a trained optical flow estimation network with the motion information provided by a traditional (non-deep) video compression scheme, e.g. H.266/VVC, as we believe the motion information of H.266/VVC achieves a better rate-distortion trade-off. In the online stage, we further optimize the latent features of the optical flows with a gradient descent-based algorithm for the video to be compressed, so as to enhance the adaptivity of the optical flows. We conduct experiments on a state-of-the-art deep video compression scheme, DCVC. Experimental results demonstrate that the proposed offline and online enhancement together achieves on average 12.8% bitrate saving on the tested videos, without increasing the model or computational complexity of the decoder side.
We address the problem of efficiently compressing video for conferencing-type applications. We build on recent approaches based on image animation, which can achieve good reconstruction quality at very low bitrate by representing face motions with a compact set of sparse keypoints. However, these methods encode video in a frame-by-frame fashion, i.e. each frame is reconstructed from a reference frame, which limits the reconstruction quality when the bandwidth is larger. Instead, we propose a predictive coding scheme which uses image animation as a predictor, and codes the residual with respect to the actual target frame. The residuals can be in turn coded in a predictive manner, thus removing efficiently temporal dependencies. Our experiments indicate a significant bitrate gain, in excess of 70% compared to the HEVC video standard and over 30% compared to VVC, on a datasetof talking-head videos