Title: DeepVCA: Deep Video Complexity Analyzer

Authors: Hadi Amirpour (AAU, Klagenfurt, Austria), Klaus Schoeffmann (AAU, Klagenfurt, Austria), Mohammad Ghanbari (University of Essex, UK), Christian Timmerer (AAU, Klagenfurt, Austria)

Abstract: Video streaming and its applications are growing rapidly, making video optimization a primary target for content providers looking to enhance their services. Enhancing the quality of videos requires the adjustment of different encoding parameters such as bitrate, resolution, and frame rate. To avoid brute force approaches for predicting optimal encoding parameters, video complexity features are typically extracted and utilized. To predict optimal encoding parameters effectively, content providers traditionally use unsupervised feature extraction methods, such as ITU-T’s Spatial Information ( SI ) and Temporal Information ( TI ) to represent the spatial and temporal complexity of video sequences. Recently, Video Complexity Analyzer (VCA) was introduced to extract DCT-based features to represent the complexity of a video sequence (or parts thereof). These unsupervised features, however, cannot accurately predict video encoding parameters. To address this issue, this paper introduces a novel supervised feature extraction method named DeepVCA, which extracts the spatial and temporal complexity of video sequences using deep neural networks. In this approach, the encoding bits required to encode each frame in intra-mode and inter-mode are used as labels for spatial and temporal complexity, respectively. Initially, we benchmark various deep neural network structures to predict spatial complexity. We then leverage the similarity of features used to predict the spatial complexity of the current frame and its previous frame to rapidly predict temporal complexity. This approach is particularly useful as the temporal complexity may depend not only on the differences between two consecutive frames but also on their spatial complexity. Our proposed approach demonstrates significant improvement over unsupervised methods, especially for temporal complexity. As an example application, we verify the effectiveness of these features in predicting the encoding bitrate and encoding time of video sequences, which are crucial tasks in video streaming. The source code and dataset are available at https://github.com/cd-athena/ DeepVCA.


Title: Cloud Storage Tier Optimization through Storage Object Classification

Authors: Akif Quddus Khan, Mihhail Matskin, Radu Prodan, Christoph Bussler, Dumitru Roman, Ahmet Soylu

Abstract: Cloud storage adoption has increased over the years given the high demand for fast processing, low access latency, and ever-increasing amount of data being generated by, e.g., Internet of Things (IoT) applications. In order to meet the users’ demands and provide a cost-effective solution, cloud service providers (CSPs) offer tiered storage; however, keeping the data in one tier is not cost-effective. In this respect, cloud storage tier optimization involves aligning data storage needs with the most suitable and cost-effective storage tier, thus reducing costs while ensuring data availability and meeting performance requirements. Ideally, this process considers the trade-off between performance and cost, as different storage tiers offer different levels of performance and durability. It also encompasses data lifecycle management, where data is automatically moved between tiers based on access patterns, which in turn impacts the storage cost. In this respect, this article explores two novel classification approaches, rule-based and game theory-based, to optimize cloud storage cost by reassigning data between different storage tiers. Four distinct storage tiers are considered: premium, hot, cold, and archive. The viability and potential of the proposed approaches are demonstrated by comparing cost savings and analyzing the computational cost using both fully-synthetic and semi-synthetic datasets with static and dynamic access patterns. The results indicate that the proposed approaches have the potential to significantly reduce cloud storage cost, while being computationally feasible for practical applications. Both approaches are lightweight and industry- and platform-independent.

Computing, https://link.springer.com/journal/607

Radu Prodan has been invited and will participate as a general chair at the ICONIC 2024, April 26-27, 2024, at Lovely Professional University, Punjab, India.

The Conference will provide a platform for scientists, researchers, academicians, industrialists, and students to assimilate the knowledge and get the opportunity to discuss and share insights through deep-dive research findings on the recent disruptions and developments in computing. All technical sessions will largely be steering Network Technologies, Artificial Intelligence and ethics, Advances in Computing, Futuristic Trends in Data Science, Security and Privacy, Data Mining and Information Retrieval.


  • To provide a platform to facilitate the exchange of knowledge, ideas, and innovations among scientists, researchers, academicians, industrialists, and students.
  • To deliberate and disseminate the recent advancements and challenges in the computing sciences.
  • To enable the delegates to establish research or business relations and find international linkage for future collaborations.

The 13th Video Browser Showdown (VBS 2024) was held on 29th January, 2024, in Amsterdam, The Netherlands, at the International Conference on Multimedia Modeling (MMM 2024). 12 international teams (from Austria, China, Czech Republic, Germany, Greece, Iceland, Ireland, Italy, Singapore, Switzerland, The Netherlands, Vietnam) competed over about 6 hours for quickly and accurately solving many search tasks of different types (known-item search/KIS, ad-hoc-video search/AVS, question-answering/QA) in three datasets with about 2500 hours of video content. Like in previous years, this large-scale international video retrieval challenge was an exciting event that demonstrated the state-of-the-art performance of interactive video retrieval systems.

ACM MMSys 2024, Bari, Italy, Apr. 15-18, 2024 

Authors: Emanuele Artioli (Alpen-Adria-Universität Klagenfurt, Austria), Farzad Tashtarian (Alpen-Adria-Universität Klagenfurt, Austria), and Christian Timmerer (Alpen-Adria-Universität Klagenfurt, Austria)

Abstract: As the popularity of video streaming entertainment continues to grow, understanding how users engage with the content and react to its changes becomes a critical success factor for every stakeholder. User engagement, i.e., the percentage of video the user watches before quitting, is central to customer loyalty, content personalization, ad relevance, and A/B testing. This paper presents DIGITWISE, a digital twin-based approach for modeling adaptive video streaming engagement. Traditional adaptive bitrate (ABR) algorithms assume that all users react similarly to video streaming artifacts and network issues, neglecting individual user sensitivities. DIGITWISE leverages the concept of a digital twin, a digital replica of a physical entity, to model user engagement based on past viewing sessions. The digital twin receives input about streaming events and utilizes supervised machine learning to predict user engagement for a given session. The system model consists of a data processing pipeline, machine learning models acting as digital twins, and a unified model to predict engagement. DIGITWISE employs the XGBoost model in both digital twins and unified models. The proposed architecture demonstrates the importance of personal user sensitivities, reducing user engagement prediction error by up to 5.8% compared to non-user-aware models. Furthermore, DIGITWISE can optimize content provisioning and delivery by identifying the features that maximize engagement, providing an average engagement increase of up to 8.6 %.

Keywords: digital twin, user engagement, xgboost



Funded by the EU, HiPEAC (High-Performance Edge And Cloud computing) is the premier focal point for networking, dissemination, training, and collaboration activities in Europe for researchers, industry, and policy related to computing systems.

The HiPEAC webinar series allows you to keep up to date on the latest advances in computer architecture and compilation research via online sessions, which can be accessed anywhere.

The Graph-Massivizer Project webinar took place on November 29, 2023. After an introduction by Nuria De LamaRadu Prodan presented the background of GraphProcessing, from Euler‘s five-node graphs to the massive graphs of today, and the motivations for the project.

 Project details have been presented by Reza Farahani and Matteo Angelinelli.

WEBINAR is online now: https://www.youtube.com/watch?v=YW7pD6nPMhs



With the current popularity of ECO in the Asia–Pacific (APAC), the Bitmovin team in APAC, led by Adrian Britton, expressed an interest in the energy-aware research initiatives conducted within the GAIA project in Austria. Following an introductory meeting between the APAC team and AAU on October 17, 2023, both teams decided to meet in person on November 21, 2023, to explore the topics further.

The meeting proved to be highly productive, centering around two recent research topics:

– VE-Match: Video Encoding Matching-Based Model in the Cloud and Edge (presented by Samira Afzal & Narges Mehran)

– Energy-aware Spatial and Temporal Resolution Selection for Per-Title (presented by Mohammad Ghasempour & Hadi Amirpour)

Many interesting Q&As appeared during each presentation due to customer and provider requirements and the future insight of climate-friendly video streaming in the Cloud and Edge. The fruitful discussions opened up avenues for future exploration in this dynamic field.

The 19th International Conference on emerging Networking EXperiments and Technologies (CoNEXT) Paris, France, December 5-8, 2023

Authors: Leonardo Peroni (IMDEA Networks Institute), Sergey Gorinsky (IMDEA Networks Institute), Farzad Tashtarian (Alpen-Adria-Universität Klagenfurt, Austria), and Christian Timmerer (Alpen-Adria-Universität Klagenfurt, Austria).

Abstract: Quality of Experience (QoE) and QoE models are of an increasing importance to networked systems. The traditional QoE modeling for video streaming applications builds a one-size-fits-all QoE model that underserves atypical viewers who perceive QoE differently. To address the problem of atypical viewers, this paper proposes iQoE (individualized QoE), a method that employs explicit, expressible, and actionable feedback from a viewer to construct a personalized QoE model for this viewer. The iterative iQoE design exercises active learning and combines a novel sampler with a modeler. The chief emphasis of our paper is on making iQoE sample-efficient and accurate.
By leveraging the Microworkers crowdsourcing platform, we conduct studies with 120 subjects who provide 14,400 individual scores. According to the subjective studies, a session of about 22 minutes empowers a viewer to construct a personalized QoE model that, compared to the best of the 10 baseline models, delivers the average accuracy improvement of at least 42% for all viewers and at least 85\% for the atypical viewers. The large-scale simulations based on a new technique of synthetic profiling expand the evaluation scope by exploring iQoE design choices, parameter sensitivity, and generalizability.


International Conference on Visual Communications and Image Processing (IEEE VCIP’23)


Authors: Vignesh V Menon, Reza Farahani, Prajit T Rajendran, Samira Afzal, Klaus Schoeffmann, Christian Timmerer

Abstract: With the emergence of multiple modern video codecs, streaming service providers are forced to encode, store, and transmit bitrate ladders of multiple codecs separately, consequently suffering from additional energy costs for encoding, storage, and transmission. To tackle this issue, we introduce an online energy-efficient Multi-Codec Bitrate ladder Estimation scheme (MCBE) for adaptive video streaming applications. In MCBE, quality representations within the bitrate ladder of new-generation codecs (e.g., HEVC, AV1) that lie below the predicted rate-distortion curve of the AVC codec are removed. Moreover, perceptual redundancy between representations of the bitrate ladders of the considered codecs is also minimized based on a Just Noticeable Difference (JND) threshold. Therefore, random forest-based models predict the VMAF of bitrate ladder representations of each codec. In a live streaming session where all clients support the decoding of AVC, HEVC, and AV1, MCBE achieves impressive results, reducing cumulative encoding energy by 56.45%, storage energy usage by 94.99%, and transmission energy usage by 77.61% (considering a JND of six VMAF points). These energy reductions are in comparison to a baseline bitrate ladder encoding based on current industry practice.

Authors: Reza Saeedinia (University of Tehran), S. Omid Fatemi (University of Tehran),Daniele Lorenzi (Alpen-Adria Universität Klagenfurt), Farzad Tashtarian (Alpen-Adria Universität Klagenfurt),  Christian Timmerer (Alpen-Adria Universität Klagenfurt)

Abstract: Live user-generated content (UGC) has increased significantly in video streaming applications. Improving the quality of experience (QoE) for users is a crucial consideration in UGC live streaming, where a user can be both a subscriber and a streamer. Resource allocation is an NP-complete task in UGC live streaming due to many subscribers and streamers with varying requests, bandwidth limitations, and network constraints. In this paper, to decrease the execution time of the resource allocation algorithm, we first process streamers’ and subscribers’ requests and then aggregate them into a limited number of groups based on their preferences. Second, we
perform resource allocation for these groups that we call communities. We formulate the resource allocation problem for communities into an optimization problem. With an efficient aggregation of subscribers and streamers at the core of the proposed architecture, the computational complexity of the optimization problem is reduced, consequently improving QoE. This improvement occurs because of the prompt reaction to the bandwidth fluctuations and, subsequently, appropriate resource allocation by the proposed model. We conduct experiments in various scenarios. The results show an average of 41% improvement in execution time. To evaluate the impact of bandwidth fluctuations on the proposed algorithm, we employ two network traces: AmazonFCC and NYUBUS. The results show 4%, and 28% QoE improvement in a scenario with 5
streamers over the AmazonFCC and the NYUBUS network traces, respectively.

Link: 13th International Conference on Computer and Knowledge Engineering (ICCKE)