The ever growing bandwidth in access networks, in combination with IPTV and video on demand (VoD) offerings, opens up unlimited possibilities to the users. The operators can no longer compete solely on the number of channels or content and increasingly make high definition channels and quality of experience (QoE) a service differentiator. Currently, the most reliable way of assessing and measuring QoE is conducting subjective experiments, where human observers evaluate a series of short video sequences, using one of the international standardized subjective quality assessment methodologies. Unfortunately, since these subjective experiments need to be conducted in controlled environments and pose limitations on the sequences and overall experiment duration they cannot be used for real-life QoE assessment of IPTV and VoD services. In this article, we propose a novel subjective quality assessment methodology based on full-length movies. Our methodology enables audiovisual quality assessment in the same environments and under the same conditions users typically watch television. Using our new methodology we conducted subjective experiments and compared the outcome with the results from a subjective test conducted using a standardized method. Our findings indicate significant differences in terms of impairment visibility and tolerance and highlight the importance of real-life QoE assessment.

Recently an impressive development in immersive technologies, such as Augmented Reality (AR), Virtual Reality (VR) and 360 video, has been witnessed. However, methods for quality assessment have not been keeping up. This paper studies quality assessment of 360 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 video. Audiovisual quality, simulator sickness symptoms, and exploration behavior were evaluated with short (from 10 seconds to 30 seconds) 360 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 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 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.

Deepfakes have raised significant concerns due to their potential to spread false information and compromise the integrity of digital media. Current deepfake detection models often struggle to generalize across a diverse range of deepfake generation techniques and video content. In this work, we propose a Generative Convolutional Vision Transformer (GenConViT) for deepfake video detection. Our model combines ConvNeXt and Swin Transformer models for feature extraction, and it utilizes an Autoencoder and Variational Autoencoder to learn from latent data distributions. By learning from the visual artifacts and latent data distribution, GenConViT achieves an improved performance in detecting a wide range of deepfake videos. The model is trained and evaluated on DFDC, FF++, TM, DeepfakeTIMIT, and Celeb-DF (v2) datasets. The proposed GenConViT model demonstrates strong performance in deepfake video detection, achieving high accuracy across the tested datasets. While our model shows promising results in deepfake video detection by leveraging visual and latent features, we demonstrate that further work is needed to improve its generalizability when encountering out-of-distribution data. Our model provides an effective solution for identifying a wide range of fake videos while preserving the integrity of media.

The scalable extension of H.264/AVC (SVC) was recently standardized, and offers scalability at a minor penalty in rate-distortion efficiency when compared to single-layer H.264/AVC coding. In SVC, a scaled version of the original video sequence can easily be extracted by dropping layers from the stream. However, most of the video content nowadays is still produced in a single-layer format. While decoding and reencoding is a possible solution to introduce scalability in the existing bitstreams, this is an approach which requires a tremendous amount of time and effort. In this paper, we show that transcoding can be used to intelligently derive scalable bitstreams from existing single-layer streams. We focus on SNR scalability, and introduce techniques that are able to create multiple quality layers in the bitstreams. We also discuss bitstream rewriting from SVC to H.264/AVC, and examine how our newly proposed architectures can benefit from the changes that were introduced for bitstream rewriting. Architectures with different rate distribution flexibility and computational complexity are discussed. Rate-distortion performance of transcoding is shown to be comparable to that of reencoding at a fraction of the time needed for the latter.