Jens Krüger

Filippo Mantovani, Fabio Banchelli, Pablo Vizcaino et al.

This paper presents EPAC, a RISC-V-based accelerator chip developed within the European Processor Initiative (EPI) as part of a multi-year, multi-partner effort to build a European HPC processor ecosystem. EPAC is implemented in GlobalFoundries 22FDX (GF22FDX) technology, covers an area of 27 sq mm with approximately 0.3 billion transistors, and integrates three distinct RISC-V compute tiles targeting different workload classes: VEC, a vector processing tile for double-precision HPC workloads; STX, a many-core tile optimized for stencil and machine learning computations; and VRP, a variable-precision tile for iterative numerical solvers requiring extended floating-point formats. All tiles are connected through a Coherent Hub Interface (CHI) based network-on-chip with a distributed L2 cache system and communicate with external memory via a SerDes link. The chip was taped out in GF22FDX technology and successfully brought up, with all major IP blocks validated. This paper describes the architecture of each tile and the uncore infrastructure, the integration and physical implementation process, and the board-level bring-up activities. It also reflects on the engineering and coordination lessons learned from a full chip design effort distributed across academic and industrial partners in Europe.

Sebastian Cmentowski, Jens Krüger

Driven by the games community, virtual reality setups have lately evolved into affordable and consumer-ready mobile headsets. However, despite these promising improvements, it remains challenging to convey immersive and engaging VR games as players are usually limited to experience the virtual world by vision and hearing only. One prominent example of such open challenges is the disparity between the real surroundings and the virtual environment. As virtual obstacles usually do not have a physical counterpart, players might walk through walls enclosing the level. Thus, past research mainly focussed on multisensory collision feedback to deter players from ignoring obstacles. However, the underlying causative reasons for such unwanted behavior have mostly remained unclear. Our work investigates how task types and wall appearances influence the players' incentives to walk through virtual walls. Therefore, we conducted a user study, confronting the participants with different task motivations and walls of varying opacity and realism. Our evaluation reveals that players generally adhere to realistic behavior, as long as the experience feels interesting and diverse. Furthermore, we found that opaque walls excel in deterring subjects from cutting short, whereas different degrees of realism had no significant influence on walking trajectories. Finally, we use collected player feedback to discuss individual reasons for the observed behavior.

Sebastian Cmentowski, Andrey Krekhov, Jens Krüger

On a journey, a backpack is a perfect place to store and organize the necessary provisions and tools. Similarly, carrying and managing items is a central part of most digital games, providing significant prospects for the player experience. Even though VR games are gradually becoming more mature, most of them still avoid this essential feature. Some of the reasons for this deficit are the additional requirements and challenges that VR imposes on developers to achieve a compelling user experience. We structure the ample design space of VR inventories by analyzing popular VR games and developing a structural taxonomy. We combine our insights with feedback from game developers to identify the essential building blocks and design choices. Finally, we propose meaningful design implications and demonstrate the practical use of our work in action.

Jonas Ney, Dominik Loroch, Vladimir Rybalkin et al.

Deep Neural Networks (DNNs) are capable of solving complex problems in domains related to embedded systems, such as image and natural language processing. To efficiently implement DNNs on a specific FPGA platform for a given cost criterion, e.g. energy efficiency, an enormous amount of design parameters has to be considered from the topology down to the final hardware implementation. Interdependencies between the different design layers have to be taken into account and explored efficiently, making it hardly possible to find optimized solutions manually. An automatic, holistic design approach can improve the quality of DNN implementations on FPGA significantly. To this end, we present a cross-layer design space exploration methodology. It comprises optimizations starting from a hardware-aware topology search for DNNs down to the final optimized implementation for a given FPGA platform. The methodology is implemented in our Holistic Auto machine Learning for FPGAs (HALF) framework, which combines an evolutionary search algorithm, various optimization steps and a library of parametrizable hardware DNN modules. HALF automates both the exploration process and the implementation of optimized solutions on a target FPGA platform for various applications. We demonstrate the performance of HALF on a medical use case for arrhythmia detection for three different design goals, i.e. low-energy, low-power and high-throughput respectively. Our FPGA implementation outperforms a TensorRT optimized model on an Nvidia Jetson platform in both throughput and energy consumption.

Sebastian Cmentowski, Andrey Krekhov, André Zenner et al.

Using virtual reality setups, users can fade out of their surroundings and dive fully into a thrilling and appealing virtual environment. The success of such immersive experiences depends heavily on natural and engaging interactions with the virtual world. As developers tend to focus on intuitive hand controls, other aspects of the broad range of full-body capabilities are easily left vacant. One repeatedly overlooked input modality is the user's gait. Even though users may walk physically to explore the environment, it usually does not matter how they move. However, gait-based interactions, using the variety of information contained in human gait, could offer interesting benefits for immersive experiences. For instance, stealth VR-games could profit from this additional range of interaction fidelity in the form of a sneaking-based input modality. In our work, we explore the potential of sneaking as a playful input modality for virtual environments. Therefore, we discuss possible sneaking-based gameplay mechanisms and develop three technical approaches, including precise foot-tracking and two abstraction levels. Our evaluation reveals the potential of sneaking-based interactions in IVEs, offering unique challenges and thrilling gameplay. For these interactions, precise tracking of individual footsteps is unnecessary, as a more abstract approach focusing on the players' intention offers the same experience while providing better comprehensible feedback. Based on these findings, we discuss the broader potential and individual strengths of our gait-centered interactions.

Sebastian Cmentowski, Jens Krüger

In early 2020, the virus SARS-CoV-2 evolved into a new pandemic, forcing governments worldwide to establish social distancing measures. Consequently, people had to switch to online media, such as social networks or videotelephony, to keep in touch with friends and family. In this context, online games, combining entertainment with social interactions, also experienced a notable growth. In our work, we focused on the potential of games as a replacement for social contacts in the COVID-19 crisis. Our online survey results indicate that the value of games for social needs depends on individual gaming habits. Participants playing mostly multiplayer games increased their playtime and mentioned social play as a key motivator. Contrarily, non-players were not motivated to add games as communication channels. We deduce that such crises mainly catalyze existing gaming habits.

Andrey Krekhov, Daniel Preuß, Sebastian Cmentowski et al.

Watching others play is a key ingredient of digital games and an important aspect of games user research. However, spectatorship is not very popular in virtual reality, as such games strongly rely on one's feelings of presence. In other words, the head-mounted display creates a barrier between the player and the audience. We contribute an alternative watching approach consisting of two major components: a dynamic view frustum that renders the game scene from the current spectator position and a one-way mirror in front of the screen. This mirror, together with our silhouetting algorithm, allows seeing the player's reflection at the correct position in the virtual world. An exploratory survey emphasizes the overall positive experience of the viewers in our setup. In particular, the participants enjoyed their ability to explore the virtual surrounding via physical repositioning and to observe the blended player during object manipulations. Apart from requesting a larger screen, the participants expressed a strong need to interact with the player. Consequently, we suggest utilizing our technology as a foundation for novel playful experiences with the overarching goal to transform the passive spectator into a collocated player.

Sebastian Cmentowski, Andrey Krekhov, Ann-Marie Müller et al.

Virtual reality (VR) games are gradually becoming more elaborated and feature-rich, but fail to reach the complexity of traditional digital games. One common feature that is used to extend and organize complex gameplay is the in-game inventory, which allows players to obtain and carry new tools and items throughout their journey. However, VR imposes additional requirements and challenges that impede the implementation of this important feature and hinder games to unleash their full potential. Our current work focuses on the design space of inventories in VR games. We introduce this sparsely researched topic by constructing a first taxonomy of the underlying design considerations and building blocks. Furthermore, we present three different inventories that were designed using our taxonomy and evaluate them in an early qualitative study. The results underline the importance of our research and reveal promising insights that show the huge potential for VR games.

Sebastian Cmentowski, Andrey Krekhov, Jens Krüger

In virtual reality games, players dive into fictional environments and can experience a compelling and immersive world. State-of-the-art VR systems allow for natural and intuitive navigation through physical walking. However, the tracking space is still limited, and viable alternatives are required to reach further virtual destinations. Our work focuses on the exploration of vast open worlds - an area where existing local navigation approaches such as the arc-based teleport are not ideally suited and world-in-miniature techniques potentially reduce presence. We present a novel alternative for open environments: Our idea is to equip players with the ability to switch from first-person to a third-person bird's eye perspective on demand. From above, players can command their avatar and initiate travels over large distance. Our evaluation reveals a significant increase in spatial orientation while avoiding cybersickness and preserving presence, enjoyment, and competence. We summarize our findings in a set of comprehensive design guidelines to help developers integrate our technique.

Andrey Krekhov, Sebastian Cmentowski, Katharina Emmerich et al.

Virtual reality setups are particularly suited to create a tight bond between players and their avatars up to a degree where we start perceiving the virtual representation as our own body. We hypothesize that such an illusion of virtual body ownership (IVBO) has a particularly high, yet overlooked potential for nonhumanoid avatars. To validate our claim, we use the example of three very different creatures---a scorpion, a rhino, and a bird---to explore possible avatar controls and game mechanics based on specific animal abilities. A quantitative evaluation underpins the high game enjoyment arising from embodying such nonhuman morphologies, including additional body parts and obtaining respective superhuman skills, which allows us to derive a set of novel design implications. Furthermore, the experiment reveals a correlation between IVBO and game enjoyment, which is a further indication that nonhumanoid creatures offer a meaningful design space for VR games worth further investigation.

Andrey Krekhov, Sebastian Cmentowski, Jens Krüger

Virtual reality offers the unique possibility to experience a virtual representation as our own body. In contrast to previous research that predominantly studied this phenomenon for humanoid avatars, our work focuses on virtual animals. In this paper, we discuss different body tracking approaches to control creatures such as spiders or bats and the respective virtual body ownership effects. Our empirical results demonstrate that virtual body ownership is also applicable for nonhumanoids and can even outperform human-like avatars in certain cases. An additional survey confirms the general interest of people in creating such experiences and allows us to initiate a broad discussion regarding the applicability of animal embodiment for educational and entertainment purposes.

Andrey Krekhov, Michael Michalski, Jens Krüger

The amount of visual communication we are facing is rapidly increasing, and skills to process, understand, and generate visual representations are in high demand. Especially students focusing on computer graphics and visualization can benefit from a more diverse education on visual literacy, as they often have to work on graphical representations for broad masses after their graduation. Our proposed teaching approach incorporates basic design thinking principles into traditional visualization and graphics education. Our course was inspired by the book Dear Data that was the subject of a lively discussion at the closing capstone of IEEE VIS 2017. The paper outlines our 12-week teaching experiment and summarizes the results extracted from accompanying questionnaires and interviews. In particular, we provide insights into the creation process and pain points of visualization novices, discuss the observed interplay between visualization tasks and design thinking, and finally draw design implications for visual literacy education in general.

Andrey Krekhov, Sebastian Cmentowski, Andre Waschk et al.

Visualizations rely on highlighting to attract and guide our attention. To make an object of interest stand out independently from a number of distractors, the underlying visual cue, e.g., color, has to be preattentive. In our prior work, we introduced Deadeye as an instantly recognizable highlighting technique that works by rendering the target object for one eye only. In contrast to prior approaches, Deadeye excels by not modifying any visual properties of the target. However, in the case of 2D visualizations, the method requires an additional setup to allow dichoptic presentation, which is a considerable drawback. As a follow-up to requests from the community, this paper explores Deadeye as a highlighting technique for 3D visualizations, because such stereoscopic scenarios support dichoptic presentation out of the box. Deadeye suppresses binocular disparities for the target object, so we cannot assume the applicability of our technique as a given fact. With this motivation, the paper presents quantitative evaluations of Deadeye in VR, including configurations with multiple heterogeneous distractors as an important robustness challenge. After confirming the preserved preattentiveness (all average accuracies above 90 %) under such real-world conditions, we explore VR volume rendering as an example application scenario for Deadeye. We depict a possible workflow for integrating our technique, conduct an exploratory survey to demonstrate benefits and limitations, and finally provide related design implications.