Generative AI has made text-guided inpainting a powerful image editing tool, but at the same time a growing challenge for media forensics. Existing benchmarks, including our text-guided inpainting forgery (TGIF) dataset, show that image forgery localization (IFL) methods can localize manipulations in spliced images but struggle in fully regenerated (FR) images, while synthetic image detection (SID) methods can detect fully regenerated images but cannot perform localization. With new generative inpainting models emerging and the open problem of localization in FR images remaining, updated datasets and benchmarks are needed. We introduce TGIF2, an extended version of TGIF, that captures recent advances in text-guided inpainting and enables a deeper analysis of forensic robustness. TGIF2 augments the original dataset with edits generated by FLUX.1 models, as well as with random non-semantic masks. Using the TGIF2 dataset, we conduct a forensic evaluation spanning IFL and SID, including fine-tuning IFL methods on FR images and generative super-resolution attacks. Our experiments show that both IFL and SID methods degrade on FLUX.1 manipulations, highlighting limited generalization. Additionally, while fine-tuning improves localization on FR images, evaluation with random non-semantic masks reveals object bias. Furthermore, generative super-resolution significantly weakens forensic traces, demonstrating that common image enhancement operations can undermine current forensic pipelines. In summary, TGIF2 provides an updated dataset and benchmark, which enables new insights into the challenges posed by modern inpainting and AI-based image enhancements. TGIF2 is available at https://github.com/IDLabMedia/tgif-dataset .

Automatic anatomical landmark localization in medical imaging requires not just accurate predictions but reliable uncertainty quantification for effective clinical decision support. Current uncertainty quantification approaches often fall short, particularly when combined with normality assumptions, systematically underestimating total predictive uncertainty. This paper introduces conformal prediction as a framework for reliable uncertainty quantification in anatomical landmark localization, addressing a critical gap in automatic landmark localization. We present two novel approaches guaranteeing finite-sample validity for multi-output prediction: multi-output regression-as-classification conformal prediction (M-R2CCP) and its variant multi-output regression to classification conformal prediction set to region (M-R2C2R). Unlike conventional methods that produce axis-aligned hyperrectangular or ellipsoidal regions, our approaches generate flexible, non-convex prediction regions that better capture the underlying uncertainty structure of landmark predictions. Through extensive empirical evaluation across multiple 2D and 3D datasets, we demonstrate that our methods consistently outperform existing multi-output conformal prediction approaches in both validity and efficiency. This work represents a significant advancement in reliable uncertainty estimation for anatomical landmark localization, providing clinicians with trustworthy confidence measures for their diagnoses. While developed for medical imaging, these methods show promise for broader applications in multi-output regression problems.

3D Gaussian Splatting (3DGS) has recently emerged as a promising contender to Neural Radiance Fields (NeRF) in 3D scene reconstruction and real-time novel view synthesis. 3DGS outperforms NeRF in training and inference speed but has substantially higher storage requirements. To remedy this downside, we propose POTR, a post-training 3DGS codec built on two novel techniques. First, POTR introduces a novel pruning approach that uses a modified 3DGS rasterizer to efficiently calculate every splat’s individual removal effect simultaneously. This technique results in 2-4× fewer splats than other post-training pruning techniques and as a result also significantly accelerates inference with experiments demonstrating 1.5-2× faster inference than other compressed models. Second, we propose a novel method to recompute lighting coefficients, significantly reducing their entropy without using any form of training. Our fast and highly parallel approach especially increases AC lighting coefficient sparsity, with experiments demonstrating increases from 70% to 97%, with minimal loss in quality. Finally, we extend POTR with a simple fine-tuning scheme to further enhance pruning, inference, and rate-distortion performance. Experiments demonstrate that POTR, even without fine-tuning, consistently outperforms all other post-training compression techniques in both rate-distortion performance and inference speed.

The demand for high-quality, low-latency video streaming is placing strain on conventional internet infrastructures. This article proposes a hybrid unicast–broadcast video delivery framework designed to address this challenge by integrating advanced 5G broadcast technologies with traditional unicast methods. By offloading popular content to a broadcast network, the approach aims to alleviate congestion and enhance overall streaming efficiency. To ensure reliable video segment delivery over the broadcast network, regardless of the physical layer, we incorporate Packet Recovery (PR) and Forward Error Correction (FEC) mechanisms. Additionally, Temporal Layer Injection (TLI) is employed to further improve video quality while maintaining reduced bandwidth requirements compared to traditional unicast-only approaches. This innovative framework leverages 5G terrestrial broadcasting within Over-the-Top (OTT) streaming environments, enabling seamless delivery of adaptive video content with sub-1-second live latency. Comprehensive experimentation and evaluation through large-scale emulation demonstrate the efficacy of this hybrid approach in meeting the evolving demands of modern multimedia delivery systems. Notably, when broadcasting the top three most commonly watched video streams, 63% of viewers no longer need to request video segments via unicast, as they are efficiently delivered over broadcast channels. This hybrid model offers significant scalability, cost reduction for an ISP, and efficiently delivers content directly to user devices without additional intermediaries, improving viewer experience through low-latency, high-quality streaming.

Real-time video encoding requires efficient and accurate quality metrics to optimize performance under strict computational and latency constraints. Traditional low-complexity metrics such as PSNR and SSIM often fall short in perceptual alignment, while accurate metrics such as VMAF are too computationally intensive for real-time deployment.We present x265-pVMAF, a low-complexity perceptual quality metric integrated into the x265 encoding loop. By leveraging machine learning and efficiently extracted encoder features, x265-pVMAF bridges the gap between computational efficiency and perceptual accuracy. It replicates VMAF predictions with a correlation of 0.99, while delivering a 37× speed-up. These results establish x265-pVMAF as a practical solution for real-time video quality assessment in next-generation encoding workflows.

In response to the growing demand for low-latency video streaming, this paper presents a demonstration of a hybrid unicast-broadcast video delivery system that combines 5G terrestrial broadcasting with over-the-top (OTT) streaming methods. The demonstration features a scalable setup with an interactive dashboard, allowing users to experiment with various configurations and observe key metrics such as bandwidth usage, packet loss, buffer size, and live latency in real-time. Key techniques include Low-Latency DASH (LL-DASH) for HTTP Adaptive Streaming (HAS), packet recovery (PR) and Forward Error Correction (FEC) for reliability, Temporal Layer Injection (TLI) for enhanced quality, and Common Media Application Format (CMAF) with Chunked Transfer Encoding (CTE) for reduced latency. The demonstration shows that this scalable hybrid approach can effectively reduce unicast bandwidth to nearly 0 Mb/s in scenarios without packet loss on the broadcast network, and achieve similar bandwidth reductions in lossy broadcast networks with appropriate Forward Error Correction (FEC) settings, while maintaining a live latency lower than 1 second. These results demonstrate the system's potential for optimizing multimedia delivery, significantly reducing unicast bandwidth while maintaining low-latency streaming.

The increasing demand for high-quality video streaming, coupled with the necessity for low-latency delivery, presents significant challenges in today's multimedia landscape. In response to these challenges, this research explores the optimization of adaptive video streaming by integrating 5G terrestrial broadcasting with over-the-top (OTT) streaming methods. A comprehensive integration of forward error correction (FEC), temporal layer injection (TLI), and broadcast techniques enhance the robustness and efficiency of content delivery over broadcast networks and reduce unicast bandwidth to zero in low loss environments. Multiple strategies are compared through an extensive emulation setup for reducing latency in the end-to-end video delivery chain to sub 3-second live latency, demonstrating the effectiveness of a hybrid unicast-broadcast approach in achieving low-latency while maintaining high-quality video streaming performance with significantly reduced bandwidth. For 62.99% of viewers, unicast bandwidth can be reduced to as low as zero when broadcasting the top 3 TV channels.

A recent research direction is focused on training Deep Neural Networks (DNNs) to replicate individual subject assessments of media quality. These DNNs are referred to as Artificial Intelligence-based Observers (AIOs). An AIO is designed to simulate, in real-time, the quality ratings of a specific individual, enabling an automatic quality assessment that accounts for subjects characteristics and preferences. Training AIOs is a promising but challenging research area due to the greater noise in individual raw opinion scores compared to the Mean Opinion Score. Effective learning from noisy labels necessitates the training of complex models on large-scale datasets. Unfortunately, this is challenging for AIOs as the media quality assessment community lacks extensive datasets that include individual opinion scores. To address the complexity of the task, we first created a dataset comprising two million samples, with synthetic labels derived from human annotation. We then trained a customized network for image quality assessment, named Multi-Distortion ResNet50 (MDResNet50), on this dataset. The weights of the MDResNet50 were subsequently utilized to initialize the learning process of each AIO, thereby avoiding the need to train a complex model from scratch on a small-scale dataset with raw individual opinion scores. Computational experiments show that our approach significantly advances the state-of-the-art in the AIO research. In particular: (i) we demonstrate through a simulation the ability of AIOs to mimic two well-known behavioral characteristics of a subject, i.e., bias and inconsistency, when scoring the media quality; (ii) we train and release DNN-based AIOs that, compared to the state-of-the-art, exhibit a higher performance with a statistical significance in assessing multiple image distortions; (iii) we train AIOs that more accurately mimic the sensitivity of real subjects to noise and color saturation and also better predict the opinion score distribution compared to the state-of-the-art AIOs.

Video streaming systems aim to provide high-quality video adapted to clients’ device and network conditions. For this purpose, adaptive streaming architectures encode video content at a variety of quality levels, organized in a bitrate ladder. However, compressing a video into multiple streams is resource-intensive, which may become especially problematic in live streaming applications with real-time demands. Therefore, this paper proposes a novel solution for fast bitrate ladder creation, and provides the requirements for implementation in the H.266/VVC standard. More specifically, the proposed method creates new intermediate Combined Streams by injecting the lowest temporal layers of a higher-quality Augmentation Stream in a lower-quality Base Stream. Since the lowest layers are used as reference by the remaining layers, this procedure indirectly increases the quality of the frames in those untouched remaining layers as well. We demonstrate that injecting more layers brings both the quality and bitrate closer to that of the Augmentation Stream. The disadvantage of the Combined Streams is that their quality fluctuates more than the quality of the source streams, and that they are compressed less efficiently, comparable to going from a slower to fast or faster preset in the VVenC encoder. Most importantly, their main advantage is that they were generated at no significant additional computational complexity. In this way, the proposed method is of great benefit when generating a bitrate ladder of video streams under constrained computational resources.

In recent years, Virtual Reality (VR) has gained attention as a tool for a plethora of applications such as first-aid, firefighting and in the automotive industry. End-user immersion is a key factor in these applications to make the experience representative for its real-life counterpart. By enhancing the traditional audiovisual cues with additional sensory inputs in terms of haptic vibro-tactile and kinesthetic feedback, this immersion can be improved. But are current haptic implementations sufficient to provide the required added value? And how do they compare to other types of VR interaction? In this paper, we present a multi-modal VR training framework able to provide subjective and objective comparisons among three different interaction options: (i) haptic gloves, (ii) traditional VR controllers, and (iii) non-haptic handtracking. We performed a user test where the different interactivity flavours were compared in terms of their influence on both subjective perception and objective performance of the end-user by means of three VR training scenarios. The subjective results show an aversion towards non-haptic handtracking for constrained, cognitively light tasks while a preference towards controllers exist for more cognitively heavy multi-tasking. This is however not reflected in objective results, where differences between interaction methods are far less pronounced.

Applications providing ultra-low-latency video streaming to large audiences require fast and efficient packet-loss repair. Previous methods utilizing keyframe injection (such as the High Efficiency Streaming Protocol) have a low impact on the repaired stream quality, but at a cost of a significant bitrate spike during repair. In this paper, we propose injecting P-frames for packet-loss repair of ultra-low-latency streaming. We implemented (open-source) and evaluated our approach in both H.265/HEVC and H.266/VVC standards. Through extensive evaluations, we demonstrate that the proposed solution significantly reduces bitrate overhead while maintaining a similar or lower decrease in quality compared to existing packet-loss-repair techniques. Overall, the proposed approach offers a promising solution to ensure reliable packet-loss recovery, efficient resource utilization, and a high-quality streaming experience.

Forensic watermarking enables the tracing of digital pirates that leak copyright-protected multimedia. To prevent a negative impact on the video quality or bit rate, rate-distortion-preserving watermarking exists, which represents a watermark as compression artifacts. However, this method has two main disadvantages; the detection has a high complexity and it is non-blind. Although a method based on perceptual hashing exists that speeds up the detection of a fallback watermarking system, it decreases its robustness. Therefore, this paper proposes a novel fast detection method that has less impact on the robustness than related work. Our method optimized NS-DCT-DST hashes for rate-distortion-preserving watermarking, which are more robust to content-preserving attacks. Moreover, a blind version is proposed which does not require the original video for hash extraction. As such, the detection is experimentally measured to be up to 5700 times faster, at the cost of a modest decrease in robustness. In fact, the proposed method shows good robustness to content-preserving recompression attacks when using hashes that are as small as 432 bytes. This is much smaller than related work at comparable performance. In conclusion, this paper enables fast adversary tracing using watermarks that do not impact the video’s compression efficiency.

Sending low-delay live video over error-prone channels comes with packet-loss-repair and random-access challenges. Existing solutions have a negative impact on end-users with reliable connections or users that do not switch channels. To minimize this impact, the keyframe-insertion technique extends a compression-efficient normal stream (NS) with a companion stream (CS) solely consisting of keyframes [1].

In six-degrees-of-freedom light-field (LF) experiences, the viewer's freedom is limited by the extent to which the plenoptic function was sampled. Existing LF datasets represent only small portions of the plenoptic function, such that they either cover a small volume, or they have limited field of view. Therefore, we propose a new LF image dataset "SILVR" that allows for six-degrees-of-freedom navigation in much larger volumes while maintaining full panoramic field of view. We rendered three different virtual scenes in various configurations, where the number of views ranges from 642 to 2226. One of these scenes (called Zen Garden) is a novel scene, and is made publicly available. We chose to position the virtual cameras closely together in large cuboid and spherical organisations (2.2m3 to 48m3), equipped with 180° fish-eye lenses. Every view is rendered to a color image and depth map of 2048px × 2048px. Additionally, we present the software used to automate the multiview rendering process, as well as a lens-reprojection tool that converts between images with panoramic or fish-eye projection to a standard rectilinear (i.e., perspective) projection. Finally, we demonstrate how the proposed dataset and software can be used to evaluate LF coding/rendering techniques (in this case for training NeRFs with instant-ngp). As such, we provide the first publicly-available LF dataset for large volumes of light with full panoramic field of view.

Low-delay live streaming applications desire fast channel switching and packet-loss repair capabilities. However, existing methods that provide these capabilities have a negative impact on the stream of steady-state users. To minimize this impact, techniques such as the High Efficiency Streaming Protocol (HESP) utilize keyframe injection. Such techniques combine compression-efficient normal streams with corresponding companion streams that are used in case of random access or packet loss. Unfortunately, because a companion stream is needed for every normal stream, the distribution cost and encoding complexity are considerable costs. Additionally, injecting a companion keyframe into a normal stream causes a bitrate spike. Therefore, this paper evaluates the impact of utilizing mixed-resolution keyframe injection in the H.266/VVC standard. By providing a single companion stream for all normal streams of a bitrate ladder, the three mentioned downsides can be mitigated. We found that injecting a lower-resolution keyframe effectively reduces the bitrate spike, at the cost of only a modest quality loss. For dynamic video content, the quality impact reduces over time, and is less perceptible than traditional packet-loss repair using frame copy. In conclusion, mixed-resolution HESP can reduce the bitrate spike and computational/distribution overhead of HESP when enabling fast channel switching or packet-loss repair.

Every route instruction type (e.g. map, symbol, photo) induces a specific cognitive load. However, when these types are used at different decision points in a building, the building configuration of these points also influences the induced cognitive load. Therefore, the process of route guidance results in an interaction between the instruction type and the decision point, which determines the induced cognitive load. One way of reducing cognitive load during route guidance is by using adaptive systems that show specific route instruction types at specific decision points. Therefore, in this VR experiment, the usability of such an adaptive indoor route guidance system is tested by tracking the wayfinding and gaze behavior of the users. First, the difference in wayfinding and gaze behavior between all route instruction types is compared. Next, the building configuration at the decision points is quantified through the architectural theory of space syntax, and the correlation with the wayfinding and gaze behavior is determined. Our findings indicate that adapting the route instruction type does make a difference for the user.

Recently an impressive development in immersive technologies, such as Augmented Reality (AR), Virtual Reality (VR) and 360${^\circ }$video, has been witnessed. However, methods for quality assessment have not been keeping up. This paper studies quality assessment of 360${^\circ }$video from the cross-lab tests (involving ten laboratories and more than 300 participants) carried out by the Immersive Media Group (IMG) of the Video Quality Experts Group (VQEG). These tests were addressed to assess and validate subjective evaluation methodologies for 360${^\circ }$video. Audiovisual quality, simulator sickness symptoms, and exploration behavior were evaluated with short (from 10 seconds to 30 seconds) 360${^\circ }$sequences. The following factors’ influences were also analyzed: assessment methodology, sequence duration, Head-Mounted Display (HMD) device, uniform and non-uniform coding degradations, and simulator sickness assessment methods. The obtained results have demonstrated the validity of Absolute Category Rating (ACR) and Degradation Category Rating (DCR) for subjective tests with 360${^\circ }$videos, the possibility of using 10-second videos (with or without audio) when addressing quality evaluation of coding artifacts, as well as any commercial HMD (satisfying minimum requirements). Also, more efficient methods than the long Simulator Sickness Questionnaire (SSQ) have been proposed to evaluate related symptoms with 360${^\circ }$videos. These results have been instrumental for the development of the ITU-T Recommendation P.919. Finally, the annotated dataset from the tests is made publicly available for the research community.

Forensic watermarking is used to track down digital pirates after they illegally redistribute video content. Although existing algorithms often resist common signal processing attacks, they are not always robust against camcording attacks. As a solution in the state of the art, registration methods are used to align the attacked video to the original one. However, watermark detection still fails when the quality is sufficiently decreased or when exposed to targeted attacks. Therefore, this paper proposes a novel fallback system that aims to detect the watermark when traditional methods fail. More concretely, we demonstrate that a primary watermark embedded by a traditional scheme indirectly creates a secondary watermark signal during video encoding. This secondary watermark consists of compression artifacts and is detected by the fallback system. Additionally, the proposed system incorporates video registration to cope with camcording attacks. The experimental results indicate that the fallback system has a striking increase in robustness compared to the existing methods. For example, the observed false-negative rate for targeted attacks improves from 100% to 0%. Moreover, the fallback is camcording resistant even when the traditional method combined with registration is not. In conclusion, the proposed system can be used as a fallback when traditional detection fails.

The omnidirectional nature of Virtual Reality (VR) content provides an immersive experience for the viewer. At the same time, VR content relies heavily on the quality of the video to deliver an immersive experience. This study investigates the effect of video quality degradation on aspects of the viewer's quality of experience (QoE) via subjective (i.e., a questionnaire) and objective (i.e., electroencephalogram) methods. We measured the viewer's experience of watching a five-minute-long 6DoF VR movie in four video quality versions. Analysis of the questionnaire data showed that subjective ratings of the video quality decreased in parallel with the degradation of the quality. Also, the lower video quality versions yielded lower sensory immersion and simulator sickness scores. Finally, analysis of the EEG data revealed significantly lower parietal and occipital alpha values for the low video quality versions of the content.

Research in light field (LF) processing has heavily increased over the last decade. This is largely driven by the desire to achieve the same level of immersion and navigational freedom for camera-captured scenes as it is currently available for CGI content. Standardization organizations such as MPEG and JPEG continue to follow conventional coding paradigms in which viewpoints are discretely represented on 2-D regular grids. These grids are then further decorrelated through hybrid DPCM/transform techniques. However, these 2-D regular grids are less suited for high-dimensional data, such as LFs. We propose a novel coding framework for higher-dimensional image modalities, called Steered Mixture-of-Experts (SMoE). Coherent areas in the higher-dimensional space are represented by single higher-dimensional entities, called kernels. These kernels hold spatially localized information about light rays at any angle arriving at a certain region. The global model consists thus of a set of kernels which define a continuous approximation of the underlying plenoptic function. We introduce the theory of SMoE and illustrate its application for 2-D images, 4-D LF images, and 5-D LF video. We also propose an efficient coding strategy to convert the model parameters into a bitstream. Even without provisions for high-frequency information, the proposed method performs comparable to the state of the art for low-to-mid range bitrates with respect to subjective visual quality of 4-D LF images. In case of 5-D LF video, we observe superior decorrelation and coding performance with coding gains of a factor of 4x in bitrate for the same quality. At least equally important is the fact that our method inherently has desired functionality for LF rendering which is lacking in other state-of-the-art techniques: (1) full zero-delay random access, (2) light-weight pixel-parallel view reconstruction, and (3) intrinsic view interpolation and super-resolution.

Typically, the measurement of the Quality of Experience for video sequences aims at a single value, in most cases the Mean Opinion Score (MOS). Predicting this value using various algorithms has been widely studied. However, deviation from the MOS is often handled as an unpredictable error. The approach in this contribution estimates intervals of video quality instead of the single valued MOS. Well-known video quality estimators are fused together to output a lower and upper border for the expected video quality, on the basis of a model derived from a well-known subjectively annotated dataset. Results on different datasets provide insight on the suitability of the well-known estimators for this particular approach.

Video watermarking is a well-established technology to help identify digital pirates when they illegally re-distribute multimedia content. In order to provide every client with a unique, watermarked video, the traditional distribution architectures separately encode each watermarked video. However, since these encodings require a high amount of computational resources, such architectures do not scale well to a large number of users. Therefore, this paper proposes a novel architecture that uses fast encoders instead of traditional, full encoders. The fast encoders re-use the coding information from a single, previously-encoded, unwatermarked video in order to speed up the encodings of the watermarked videos. As a result, the complexity of a fast encoder is only a fraction of the complexity of a full encoder. Due to a high correlation of the re-used coding information with the optimal coding information, the compression efficiency and watermark robustness decrease only slightly. Most importantly, the proposed fast encoder speeds up the compression process with a factor of 115, resulting in a low complexity similar to that of a video decoder. Consequently, video distributors can use the proposed architecture to deliver high-quality watermarked videos on a large-scale without requiring an excessive amount of computational resources.

The training and performance analysis of objective video quality assessment algorithms is complex due to the huge variety of possible content classes and transmission distortions. Several secondary issues such as free parameters in machine learning algorithms and alignment of subjective datasets put an additional burden on the developer. In this paper, three subsequent steps are presented to address such issues. First, the content and coding parameter space of a large-scale database is used to select dedicated subsets for training objective algorithms. This aims at providing a method for selecting the most significant contents and coding parameters from all imaginable combinations. In the practical case where only a limited set is available, it also helps us to avoid redundancy in the training subset selection. The second step is a discussion on performance measures for algorithms that employ machine-learning methods. The particularity of the performance measures is that the quality of the training and verification datasets is taken into consideration. Common issues that often use existing measures are presented, and improved or complementary methods are proposed. The measures are applied to two examples of no-reference objective assessment algorithms using the aforementioned subsets of the large-scale database. While limited in terms of practical applications, this sandbox approach of objectively predicting an objectively evaluated video sequences allows for eliminating additional influence factors from subjective studies. In the third step, the proposed performance measures are applied to the practical case of training and analyzing assessment algorithms on readily available subjectively annotated image datasets. The presentation method in this part of the paper can also be used as an exemplified recommendation for reporting in-depth information on the performance. Using this presentation method, future publications presenting newly developed quality assessment algorithms may be significantly improved.

Steered Mixture-of-Experts (SMoE) is a novel framework for the approximation, coding, and description of image modalities such as light field images and video. The future goal is to arrive at a representation for Six Degrees-of-Freedom (6DoF) image data. Previous research has shown the feasibility of real-time pixel-parallel rendering of static light field images. Each pixel is independently reconstructed by kernels that lay in its vicinity. The number of kernels involved forms the bottleneck on the achievable framerate. The goal of this paper is twofold. Firstly, we introduce pixel-level rendering of light field video, as previous work only rendered static content. Secondly, we investigate rendering using a predefined number of most significant kernels. As such, we can deliver hard real-time constraints by trading off the reconstruction quality.

Steered Mixture-of-Experts (SMoE) is a novel framework for the approximation, coding, and description of image modalities. The future goal is to arrive at a representation for Six Degrees-of-Freedom (6DoF) image data. The goal of this paper is to introduce SMoE for 4D light field videos by including the temporal dimension. However, these videos contain vast amounts of samples due to the large number of views per frame. Previous work on static light field images mitigated the problem by hard subdividing the modeling problem. However, such a hard subdivision introduces visually disturbing block artifacts on moving objects in dynamic image data. We propose a novel modeling method that does not result in block artifacts while minimizing the computational complexity and which allows for a varying spread of kernels in the spatio-temporal domain. Experiments validate that we can progressively model light field videos with increasing objective quality up to 0.97 SSIM.

High Dynamic Range (HDR) imaging captures a wide range of luminance existing in real-world scenes. Due to large luminance levels and higher brightness of HDR displays, artefacts can be more noticeable to the Human Visual System (HVS). In a first attempt to experimentally quantify those noticeable levels for HDR images, we pioneered in conducting an exhaustive and comprehensive Just Noticeable Difference (JND) subjective experiment of which the outcome is presented in this paper. Six distortions including JPEG, JPEG2000, noise, blur, contrast change, and quantization artefacts have been considered in the test. The distortions were applied to 10 HDR images in 100 distortion levels resulting a database of 6000 HDR test images. The subjects were asked to find the image JND location on each set of 100 images they had the freedom to explore. The effect of content features on the noticeable threshold selection is investigated per distortion type. Our results in some cases show a significant correlation between content features and JNDs. We are hoping that our results can contribute to further exploitation of a precise HVS model for HDR quality assessment and optimization of the coding and bit allocation in HDR compression.

Motivation: The past decade has seen the introduction of new technologies that significantly lowered the cost of genome sequencing. As a result, the amount of genomic data that must be stored and transmitted is increasing exponentially. To mitigate storage and transmission issues, we introduce a framework for lossless compression of quality scores. Results: This article proposes AQUa, an adaptive framework for lossless compression of quality scores. To compress these quality scores, AQUa makes use of a configurable set of coding tools, extended with a Context-Adaptive Binary Arithmetic Coding scheme. When benchmarking AQUa against generic single-pass compressors, file sizes are reduced by up to 38.49% when comparing with GNU Gzip and by up to 6.48% when comparing with 7-Zip at the Ultra Setting, while still providing support for random access. When comparing AQUa with the purpose-built, single-pass, and state-of-the-art compressor SCALCE, which does not support random access, file sizes are reduced by up to 21.14%. When comparing AQUa with the purpose-built, dual-pass, and state-of-the-art compressor QVZ, which does not support random access, file sizes are larger by 6.42-33.47%. However, for one test file, the file size is 0.38% smaller, illustrating the strength of our single-pass compression framework. This work has been spurred by the current activity on genomic information representation (MPEG-G) within the ISO/IEC SC29/WG11 technical committee. Availability and implementation: The software is available on Github: https://github.com/tparidae/AQUa. Contact: [email protected].

We propose a novel approach for modeling and coding color in images and video. Luminance is linearly correlated with chrominance locally, as such we can predict color given the luma value. Using the Steered Mixture-of-Experts (SMoE) approach, the image is viewed as a stochastic process over 5 random variables including the 2-D pixel locations, 1 luminance and 2 chrominance values. We model this process as a continuous joint density function by fitting a K-modal 5-D Gaussian Mixture Model (GMM). As such, the chroma values are predicted as the expectation of the conditional density. To validate, the technique was integrated within JPEG showing PSNR gains in the lower bitrate regions. A deeper analysis of the tolerance of the activation function is given through recycling color models in video sequences, yielding a high quality reconstruction over a considerable range of frames.

The proposed framework, called Steered Mixture-of-Experts (SMoE), enables a multitude of processing tasks on light fields using a single unified Bayesian model. The underlying assumption is that light field rays are instantiations of a non-linear or non-stationary random process that can be modeled by piecewise stationary processes in the spatial domain. As such, it is modeled as a space-continuous Gaussian Mixture Model. Consequently, the model takes into account different regions of the scene, their edges, and their development along the spatial and disparity dimensions. Applications presented include light field coding, depth estimation, edge detection, segmentation, and view interpolation. The representation is compact, which allows for very efficient compression yielding state-of-the-art coding results for low bit-rates. Furthermore, due to the statistical representation, a vast amount of information can be queried from the model even without having to analyze the pixel values. This allows for “blind” light field processing and classification.

MOTIVATION: The past decade has seen the introduction of new technologies that lowered the cost of genomic sequencing increasingly. We can even observe that the cost of sequencing is dropping significantly faster than the cost of storage and transmission. The latter motivates a need for continuous improvements in the area of genomic data compression, not only at the level of effectiveness (compression rate), but also at the level of functionality (e.g. random access), configurability (effectiveness versus complexity, coding tool set …) and versatility (support for both sequenced reads and assembled sequences). In that regard, we can point out that current approaches mostly do not support random access, requiring full files to be transmitted, and that current approaches are restricted to either read or sequence compression. RESULTS: We propose AFRESh, an adaptive framework for no-reference compression of genomic data with random access functionality, targeting the effective representation of the raw genomic symbol streams of both reads and assembled sequences. AFRESh makes use of a configurable set of prediction and encoding tools, extended by a Context-Adaptive Binary Arithmetic Coding scheme (CABAC), to compress raw genetic codes. To the best of our knowledge, our paper is the first to describe an effective implementation CABAC outside of its' original application. By applying CABAC, the compression effectiveness improves by up to 19% for assembled sequences and up to 62% for reads. By applying AFRESh to the genomic symbols of the MPEG genomic compression test set for reads, a compression gain is achieved of up to 51% compared to SCALCE, 42% compared to LFQC and 44% compared to ORCOM. When comparing to generic compression approaches, a compression gain is achieved of up to 41% compared to GNU Gzip and 22% compared to 7-Zip at the Ultra setting. Additionaly, when compressing assembled sequences of the Human Genome, a compression gain is achieved up to 34% compared to GNU Gzip and 16% compared to 7-Zip at the Ultra setting. AVAILABILITY AND IMPLEMENTATION: A Windows executable version can be downloaded at https://github.com/tparidae/AFresh . CONTACT: [email protected].

Video-streaming events such as virtual classrooms and video conferences require a low delay between sender and receiver. In order to achieve this requirement, and to make full use of the bandwidth capacity of each receiver, each client can be provided with a personalized bitstream of which the bit rate is continuously adapted to his current network bandwidth capacity. However, such an approach requires an excessive amount of computationally complex video encoders. Therefore, this paper proposes an architecture based on coding information calculation (CIC) modules and residual encoder (RE) modules. The CIC modules calculate coding information for the video at certain bit rates whereas the RE modules use this information to skip all encoding steps of a traditional encoder, except for the encoding of the residual. By reducing the amount of bits used to encode the residual, the RE modules can then provide bitstreams with personalized bit rates for several users at the same time. Each CIC module has approximately the same computational complexity as a traditional encoder, whereas an RE module has the approximate complexity of a decoder. The proposed architecture was evaluated for the high efficiency video coding standard, showing that the system achieves its goal of drastically reducing the computational complexity of low-delay live-streaming with many participants and suggesting that using less than six CIC modules results in the best tradeoff between compression efficiency and computational complexity.

Complex software appliances typically consist of multiple software processes running concurrently to exploit the available computational resources in the hardware. However, the computational complexity of these software processes is often variable and the processes can interfere with each other. This can be an issue for real-time applications with a fixed deadline like low delay video encoding. In the context of High Efficiency Video Coding (HEVC), a limited number of publications have focused on controlling the complexity of an HEVC video encoder. In this paper, we propose a technique to control complexity by deciding between 2Nx2N merge mode and full encoding, at different Coding Unit (CU) depths. Our results demonstrate fast convergence to a given complexity threshold after a maximum of 10 frames, and a limited loss in rate-distortion performance (on average 2.84% Bjontegaard delta rate for 60% complexity reduction).

Our challenge is the design of a “universal” bit-efficient image compression approach. The prime goal is to allow reconstruction of images with high quality. In addition, we attempt to design the coder and decoder “universal”, such that MPEG-7-like low-and mid-level descriptors are an integral part of the coded representation. To this end, we introduce a sparse Mixture-of-Experts regression approach for coding images in the pixel domain. The underlying stochastic process of the pixel amplitudes are modelled as a 3-dimensional and multi-modal Mixture-of-Gaussians with K modes. This closed form continuous analytical model is estimated using the Expectation-Maximization algorithm and describes segments of pixels by local 3-D Gaussian steering kernels with global support. As such, each component in the mixture of experts steers along the direction of highest correlation. The conditional density then serves as the regression function. Experiments show that a considerable compression gain is achievable compared to JPEG for low bitrates for a large class of images, while forming attractive low-level descriptors for the image, such as the local segmentation boundaries, direction of intensity flow and the distribution of these parameters over the image.

Complex software appliances can be deployed on hardware with limited available computational resources. This computational boundary puts an additional constraint on software applications. This can be an issue for real-time applications with a fixed time constraint such as low delay video encoding. In the context of High Efficiency Video Coding (HEVC), a limited number of publications have focused on controlling the complexity of an HEVC video encoder. In this paper, a technique is proposed to control complexity by deciding between 2N×2N merge mode and full encoding, at different Coding Unit (CU) depths. The technique is demonstrated in two encoders. The results demonstrate fast convergence to a given complexity threshold, and a limited loss in rate-distortion performance (on average 2.84% Bjontegaard delta rate for 40% complexity reduction).

In many application scenarios, video quality assessment is required to be fast and reasonably accurate. The characterization of objective algorithms by subjective assessment is well established but limited due to the small number of test samples. Verification using large-scale objectively annotated databases provides a complementary solution. In this contribution, three simple but fast measures are compared regarding their agreement on a large-scale database. In contrast to subjective experiments, not only sequence-wise but also framewise agreement can be analyzed. Insight is gained into the behavior of the measures with respect to 5952 different coding configurations of High Efficiency Video Coding (HEVC). Consistency within a video sequence is analyzed as well as across video sequences. The results show that the occurrence of discrepancies depends mostly on the configured coding structure and the source content. The detailed observations stimulate questions on the combined usage of several video quality measures for encoder optimization.

High-dynamic-range (HDR) technology allows capturing of video content at a wider range of luminance than low-dynamic-range (LDR) video. The resulting video more closely resembles the scene as perceived by the human eye. However, displays currently support only a limited range of HDR. Therefore, both an HDR version and LDR version of a video should be encoded during content acquisition. This means that the cost of encoder hardware in cameras would double. As a solution, this paper proposes a multistream video encoder that allows generating an HDR and LDR version of the same HDR video footage at roughly the same computational complexity as a single encoder, effectively allowing encoding of two dynamic-range versions of the video with a negligible increase in cost. For the LDR version, this multistream encoder results in a bit rate overhead of only 11.6% for the same quality as a two-encoder solution.

The performance of objective video quality measures is usually identified by comparing their predictions to subjective assessment results which are regarded as the ground truth. In this work we propose a complementary approach for this performance evaluation by means of a large-scale database of test sequences evaluated with several objective measurement algorithms. Such an approach is expected to detect performance anomalies that could highlight shortcomings in current objective measurement algorithms. Using realistic coding and network transmission conditions, we investigate the consistency of the prediction of different measures as well as how much their behavior can be predicted by content, coding and transmission features, discussing unexpected and peculiar behaviors, and highlighting how a large-scale database can help in identifying anomalies not easily found by means of subjective testing. We expect that this analysis will shed light on directions to pursue in order to overcome some of the limitations of existing reliability assessment methods for objective video quality measures.

With the introduction of 4K UHD video and display resolution, questions arise on the perceptual differences between 4K UHD and upsampled HD video content. In this paper, a striped pair comparison has been performed on a diverse set of 4K UHD video sources. The goal was to subjectively assess the perceived sharpness of 4K UHD and downscaled/upscaled HD video. A striped pair comparison has been applied in order to make the test as straightforward as possible for a non-expert participant population. Under these conditions and over this set of sequences, on average, on 54.8% of the sequences (17 out of 31), 4K UHD resolution content could be identified as being sharper compared to its HD down and upsampled alternative. The probabilities in which 4K UHD could be differentiated from downscaled/upscaled HD range from 83.3% for the easiest to assess sequence down to 39.7% for the most difficult sequence. Although significance tests demonstrate there is a positive sharpness difference from camera quality 4K UHD content compared to the HD downscaled/upscaled variations, it is very content dependent and all circumstances have been chosen in favor of the 4K UHD representation. The results of this test can contribute to the research process of developing metrics indicating visibility of high resolution features within specific content.

In professional markets, there is a growing need to incorporate off-the-shelf consumer-level hardware to reduce costs and speed up development. Typically, these hardware components contain hardware-accelerated video encoders, which are limited to the more popular YUV4:2:0 profiles. For screen content coding (SCC) (e.g., desktop sharing, supervisory control and data acquisition applications, etc.), chroma subsampling can produce specific artifacts that are disturbing in high-quality professional applications. Current research on SCC focuses on developing specific screen content profiles and coding tools, which are not likely to be integrated in consumer electronics. In this paper, a method is explored to improve the chroma quality of SCC using YUV4:2:0 codecs. It is possible to recover parts of the chroma component based on the luma component and the subsampling filter that is used at the encoder device. A 1.63-dB peak signal-to-noise ratio-chroma improvement and 0.051 structural similarity-chroma improvement are measured on average for screen content sequences. The proposed method still provides improvements if compression is used.

High efficiency video coding (HEVC) shows a significant advance in compression efficiency and is considered to be the successor of H.264/AVC. To incorporate the HEVC standard into real-life network applications and a diversity of other applications, efficient bit rate adaptation (transrating) algorithms are required. A current problem of transrating for HEVC is the high computational complexity associated with the encoder part of such a cascaded pixel domain transcoder. This paper focuses on deriving an optimal strategy for reducing the transcoding complexity with a complexity-scalable scheme. We propose different transcoding techniques which are able to reduce the transcoding complexity in both CU and PU optimization levels. At the CU level, CUs can be evaluated in top-to-bottom or bottom-to-top flows, in which the coding information of the input video stream is utilized to reduce the number of evaluations or to early terminate certain evaluations. At the PU level, the PU candidates are adaptively selected based on the probability of PU sizes and the co-located input PU partitioning. Moreover, with the use of different proposed methods, a complexity-scalable transrating scheme can be achieved. Furthermore, the transcoding complexity can be effectively controlled by the machine learning based approach. Simulations show that the proposed techniques provide a superior transcoding performance compared to the state-of-the-art related works. Additionally, the proposed methods can achieve a range of trade-offs between transrating complexity and coding performance. From the proposed schemes, the fastest approach is able to reduce the complexity by 82% while keeping the bitrate loss below 3%.

Communication using internet and digital media is more and more popular. Therefore, the security and privacy of data transmission are highly demanded. One effective technique providing this requirement is information hiding. This technique allows to conceal secret information into a video file, an audio, or a picture. In this paper, we propose a low complexity out-of-the-loop information hiding algorithm for a video pre-encoded with the high efficiency video coding standard. Only selected components such as the motion vector difference and transform coefficients of the video are extracted and modified, bypassing the need of fully decoding and re-encoding the video. In order to reduce the propagation error caused by hiding information, the dependency between video frames is taken into account when distributing the information over the frame. Several embedding strategies are investigated. The experimental results show that the information should be hidden in smaller blocks to reduce quality loss. Using a smart distribution of information across the frames can keep the quality loss under 1 dB PSNR for an information payload of 15 kbps. When such a strategy is used, embedding information in the transform coefficients only slightly outperforms the modification of motion vector differences.

Content providers create different versions of a video to accommodate different end-user devices and network conditions. However, each of these versions requires a resource intensive encoding process. To reduce the computational complexity of the encodings, this paper proposes a fast simultaneous encoder. This encoder takes a single video as input and creates a number of bit streams encoded with different parameters. Only one version of the video is created with a full encode, whereas encoding of the other versions is accelerated by exploiting the correlation with the fully encoded version using machine learning techniques. In a practical scenario, the fast simultaneous encoder achieves a complexity reduction of 67.3% with a bit rate increase of 5.2% compared to performing a full encode of each version.

The rapid emergence and rise of a wide range of electronic devices has led to the need for providing very different levels of video transmission services. The capabilities and performance of these devices determine the type of video streams that they are able to decode. As a way to fulfil their requirements, this paper presents a heterogeneous simultaneous encoding framework of H.264/Advanced Video Coding (AVC) and High Efficiency Video Coding (HEVC) that shares information between encoders in order to reduce the overall encoding time. In this regard, the proposed approach utilizes the H.264/AVC motion vectors of the 16x16 pixels macroblocks as predictors for the HEVC prediction units. As a consequence, the size of the motion estimation search area can be significantly reduced. Results show that an encoding time reduction of 8.95% can be achieved with negligible losses in terms of rate-distortion.

The visualization of (ultra) high-resolution compositions created from multiple input bitstreams requires several decoders at the receiving device. Therefore, not all devices can properly display such compositions. To address this problem, the input streams are decoded, merged into a single video, and re-encoded by a transcoder in the network. However, this approach requires a computationally complex re-encoding step. To reduce this complexity, information from the input bit-streams can be reused during transcoding. In HEVC, simply reusing the original encoding information is not compression efficient if the inserted content is not aligned with the grid of coded blocks. In this paper, we applied feature selection based on a decision tree, which was used in a fast HEVC transcoding model for misaligned content. The performance varies depending on the shift and average transform size in the original sequence, resulting in complexity reductions of up to 76%.

Transrating is a useful tool for adapting the bitrate of a video stream. Reducing the complexity of transrating is extremely important, especially for high efficiency video coding (HEVC) based transrating where the encoder is very time consuming. TZSearch is currently the default integer motion estimation (ME) algorithm in the HEVC reference encoder thanks to its excellent performance in reducing the complexity of ME. However, this algorithm is characterized by a fixed search area and search pattern, which can be considered sub optimal for transrating. In this paper, we improve the performance of HEVC transrating by optimizing this TZSearch algorithm. Utilizing the correlation between input and output motion vectors, we propose a fast search scheme including three steps. First, the initial search point is selected. Then, using the rate-distortion cost of this starting point, the search size is determined according to an online-trained Bayes decision rule. Finally, two proposed search algorithms are described for refining the starting point. Experimental results show that our proposed TZSearch scheme can reduce the complexity of transrating while improving the coding performance in terms of bit rate saving.

The capabilities of off-the-shelf consumer electronics are continuously increasing, making them interesting candidates to integrate into industrial visualization systems for reducing the cost of these systems or speed up development. However, in the context of video compression, hardware acceleration on consumer electronics is typically only provided for more common, consumer oriented YUV 4:2:0 profiles. Meanwhile, industrial applications often require YUV 4:4:4 formats due to synthetic visual data. Upscaling YUV 4:2:0 to YUV 4:4:4 using conventional filters results in visual artifacts, specifically for synthetic content. To improve the quality of the chroma components, this paper proposes to extend the well-known guided image filter through content-adaptive selection of the filter radius, and analyze different strategies for limiting complexity. This results in a 3.5 dB improvement of the chroma PSNR compared to conventional filters.

In an IPTV environment, different techniques exist to provide faster random access or equivalently, a faster channel switching experience, but none of them provide backward compatibility and limited overhead at the same time. In this paper, a technique is proposed to increase the random access frequency for the High Efficiency Video Coding (HEVC) standard by using a single-loop scalable version of it. This is achieved by encoding an enhancement layer with a higher frequency of random access points compared to the base layer. With this scalable configuration, a backward compatible base stream remains present. The proposed technique requires 12.4% less bitrate compared to sending the fast and slow switching streams in a non-scalable way. Compared to the almost standardized multi-loop scalable extension of HEVC, 5.5% less bitrate is needed on the core of the IPTV network and 12.5% less bitrate on the access network during steady state conditions. Moreover, the multi-loop scalable extension does not provide backward compatibility with legacy HEVC decoders.

One of the drawbacks of current 3DTV systems are the glasses required to separate the 3D views. This problem can be solved by using autostereoscopic 3D displays, which generates a 3D scene from different viewpoints. These viewpoints are synthesized in the display based on depth information of the scene. However, currently no depth information is available for stereoscopic video. Therefore, depth estimation based on the stereoscopic input is required. This depth estimation should be low-complex and real-time for consumer electronic devices. Due to this low-complexity requirement, the quality of such depth maps will be limited. In this paper, we propose a guided depth filter, which filters the estimated depth map by using the downsampled chroma signal of the corresponding texture view as a guide. This allows to refine the borders of objects which reduces visible inpainting artifacts caused by the view synthesis. The proposed algorithm has been evaluated by expert viewing to show that it is independent of the depth map estimation and view synthesis of the autostereoscopic display.

To incorporate the newly developed High Efficiency Video Coding (HEVC) standard in real-life network applications, efficient transrating algorithms are required. We propose a fast transrating scheme, based on the early prediction of the partition split-flags in P pictures. Using machine learning techniques, the correlation between co-located partitions at different quantizations is investigated. This results in a model which predicts the split-flag and gives the associated prediction accuracy so that the splitting process in the transcoder is optimized. At each partition depth, the model indicates whether the full rate-distortion cost evaluations should be performed at the current depth, or if the partition can be split immediately. Experimental results show that the proposed transcoder reduces the complexity of the transrating process by 76.04%, while maintaining the coding efficiency of a cascaded decoder-encoder.

When middlebox devices should be able to adapt an encrypted video stream in the network without having the decryption key, format-compliant partial encryption schemes should be applied. In this paper, we propose such encryption schemes for the recently standardized High Efficiency Video Coding (HEVC) standard. By encrypting specific syntax elements like the sign of the residual information, the sign of the motion vector (MV) difference, the MV prediction index, and the MV reference index, format compliance and the possibility for adaptation are offered. Scrambling performance gradually increases when shifting from encrypting the motion information to encrypting the residual sign and finally to the combination thereof. Applying all these techniques has a negligible impact on the compression efficiency.

H.264/AVC applies an adaptive in-loop deblocking filter in order to remove blocking artifacts. The deblocking filter is adaptively controlled by using the boundary strength (BS) parameter. Although the filter achieves advantages in reducing blocking artifacts, there are two problems we should consider to improve the filtering performance. The first problem is that the filter applies just a strong and the strongest filter to filter intra coded blocks. The strong filter can remove blocking artifacts at smooth regions; however, it removes detail or sharpness of pictures at complex regions. The second problem is that H.264/AVC does not pay attention in removing corner outliers which appear when an edge of an object crosses a corner of a block. Therefore, corner outliers clearly appear and reduce the subjective quality of reconstructed pictures. In this paper, we propose an adaptive deblocking filter scheme which removes corner outliers, and which utilizes the intra prediction information of intra coded blocks to adaptively select the BS. Experimental results show that our proposed filter significantly enhances the subjective quality by removing blocking artifacts at smooth regions while maintaining details or sharpness at complex regions. In addition, the objective quality is improved, resulting in PSNR gains of up to 0.16 dB.

When a video delivery system provides different representations of the same sequence, solutions can be found in simulcast coding, scalable coding, and transcoding. The major downside of transcoding is the additional complexity needed to re-encode the video sequence in its new form. When one fidelity version of the video stream should be transmitted over the network, scalable coding is less efficient compared to single layer coding because of the layering overhead. Finally, simulcast encoding results in large storage requirements. In this paper, we propose an intermediate solution providing transcoding at a low complexity by the aid of control streams. We define a control stream as a regular video stream from which residual information is removed. With these control streams, the complexity of the encoding step in the transcoder can be reduced to decoder complexity. As a result of the removal of residual information, these control streams take up 62.8% less bitrate compared to simulcast coding for a test set based on High Efficiency Video Coding (HEVC). Additionally, when compared to scalable coding, an efficient single layer video stream can be provided without the 16.6% bitrate increase caused by transmitting a layered bitstream.

Scalable video coding performance largely depends on the underlying single layer coding efficiency. In this paper, the quality scalability capabilities are evaluated on a base of the new High Efficiency Video Coding (HEVC) standard under development. To enable the evaluation, a multi-loop codec has been designed using HEVC. Adaptive inter-layer prediction is realized by including the lower layer in the reference list of the enhancement layer. As a result, adaptive scalability on frame level and on prediction unit level is accomplished. Compared to single layer coding, 19.4% Bjontegaard Delta bitrate increase is measured over approximately a 30dB to 40dB PSNR range. When compared to simulcast, 20.6% bitrate reduction can be achieved. Under equivalent conditions, the presented technique achieves 43.8% bitrate reduction over Coarse Grain Scalability of the SVC - H.264/AVC-based standard.

Scalable video coding (SVC) allows encoded bitstreams to be adapted. However, most bitstreams do not incorporate this scalability so bitstreams have to be adapted multiple times to accommodate for varying network conditions or end-user devices. Each adaptation incorporates an additional loss of quality due to transcoding. To overcome this issue, we propose a single transcoding step from H.264/AVC to SVC. Doing so, the resulting bitstream can be freely adapted without any additional quality reduction. Open-loop transcoding architectures can be used for H.264/AVC-to-SVC transcoding with a low complexity, although these architectures suffer from drift artifacts. Closed-loop transcoding, on the other hand, requires a higher complexity. To overcome the drawbacks of both systems, we propose combining both techniques.

In the current development of HEVC, compression performance improved significantly compared to H.264/AVC for both inter pictures and intra pictures. With intra compression, the main reason for this improvement is the large in crease in intra prediction directions (up to 34). The downside of having a larger number of modes is that they increase the signaling overhead in the bitstream. In this paper, a low complexity intra mode prediction algorithm is proposed which improves the mode prediction accuracy. This is achieved by exploiting the correlation between the prediction directions of the neighboring prediction units and that of the encoded prediction unit. As a result, more efficient intra mode signaling can be achieved with minimal impact on encoder and decoder complexity. On average, 0.33% bitrate improvement is obtained by employing the proposed algorithm. For sequences that are encoded with a high number of directional intra modes, around 1% bitrate improvement is measured.

An IPTV network contains characteristics of both a unicast environment and a broadcast environment. For both environments, techniques to accelerate channel switching exist. In a unicast environment a bandwidth efficient implementation to accelerate channel switching can be obtained with non scalable video coding, while for a broadcast environment the same acceleration is more optimal using scalable video coding. For a combination of both environments, as is the case for an IPTV environment, none of the solutions is optimal. In this paper, we propose a compromise between scalable and non scalable video compression to adapt to the properties of an IPTV environment. One of the proposed configurations obtains bandwidth reductions on the access network of the IPTV network between 2.4% and 4.3% with a small increase between 0.1% and 1.6% on the core network of the IPTV network.