Events

Talk BA-Talk: Realistische blickabhängige Bewegungsunschärfe zur Cybersickness-Reduktion in 360°-Virtual-Reality-Videos

28.09.2026 15:00
G30

Speaker(s): Kuvvet Kaan Urtkin

Virtual Reality ermöglicht es, digitale oder aufgezeichnete Umgebungen immersiv darzustellen. 360°-VR-Inhalte können bei der Nutzung Unwohlsein und Beschwerden hervorrufen, die unter dem Begriff Cybersickness zusammengefasst werden. Visuelle Gegenmaßnahmen wie Bewegungsunschärfe sollen solche Beschwerden verringern, können dabei aber auch die Wahrnehmung relevanter Bildinhalte beeinträchtigen. Diese Arbeit untersucht daher einen blickabhängigen Ansatz, bei dem neben der Bewegung des dargestellten Inhalts auch Kopf- und Blickbewegungen berücksichtigt werden. Dadurch soll ein mit den Augen verfolgtes Objekt weniger stark verwischt werden, wenn seine Bewegung relativ zum Blick geringer ist.

Hierfür wurde ein zeitlich normiertes Bewegungsmodell entwickelt und als Echtzeit-Prototyp in Unity umgesetzt. Das Verfahren führt die unterschiedlichen Bewegungsanteile zu einer blickrelativen Bewegung zusammen und rekonstruiert daraus eine gerichtete Bewegungsunschärfe für 360°-VR-Inhalte.

Talk MA-Talk: An Investigation of using 3D Gaussian Splatting based Scene Model for 3D Path Planning in Active Scene Reconstruction

28.09.2026 13:15
G30

Jiahao, Chen

3D Gaussian Splatting (3DGS) has recently emerged as an effective representation for high-quality 3D scene reconstruction and is increasingly applied to robotic tasks such as mapping, localization, and active scene reconstruction. Unlike implicit radiance fields, 3DGS represents scenes using explicit Gaussian primitives with spatial positions, scales, and orientations, enabling direct use of the reconstructed representation for spatial reasoning. In active scene reconstruction, robots must continuously perform 3D path planning based on the current scene model, while conventional approaches  typically require additional occupancy maps, TSDFs, or ESDFs. When a system already maintains a 3DGS scene model for its tasks, directly exploiting its geometric information can reduce the need for constructing separate planning representations and their associated computational cost.

This thesis systematically investigates the feasibility of using a 3DGS scene model for 3D path planning and introduces two novel approaches, designated GSPAR. GSPAR-Direct directly uses the positions, scales, orientations, and opacity values of anisotropic Gaussians to construct continuous collision constraints, followed by global path search and continuous path optimization in the original Gaussian scene model. GSPAR-Global instead processes and converts the Gaussian scene model into a compact 3D occupancy map through Geometry Filtering, spatial support constraints, thin surface processing, and occupancy evidence accumulation. The resulting representation supports reusable global path planning and local trajectory optimization. In addition, GSPAR-Global has been integrated with an online approach, thus validating the proposed methods for application to a continuously updated scene representation in real-world contexts.

The proposed approaches are evaluated in terms of mapping quality, representation scale, computational efficiency, and path planning performance, with comparisons against a conventional distance-field pipeline and an existing Gaussian-based planning method. The results indicate that, although the geometric accuracy of standard 3DGS is constrained by its visual reconstruction objective, the encoded spatial information remains sufficient to support effective 3D path planning. These findings demonstrate that 3DGS can serve not only as a high-quality visual representation, but also as a practical geometric representation for 3D path planning.

Talk BA-Talk: Designing and Implementing an Attention Monitoring and Feedback System for an Immersive Learning Space for ADHD

03.09.2026 14:00
G30

Speaker(s): Valentin Laubsch

Attention-deficit/hyperactivity disorder (ADHD) can make self-directed university study challenging because learners must organize their work, sustain attention, and manage distractions. Virtual reality can provide a controlled and configurable workspace, but an immersive environment alone neither observes visual behavior nor reacts to prolonged or repeated gaze away from the learning area.

This thesis extends an existing VR learning environment with configurable gaze monitoring and attention guidance. Eye-tracking data from the Meta Quest Pro are mapped to semantic focus, temporarily allowed, break, and distractor areas. Temporal and visit-based rules distinguish tolerated gaze diversions from wandering and sustained distraction. When guidance is recommended, configurable combinations of a desk edge highlight, desk focus light, room dimming, audio reduction, and a directional wave cue are selected according to the observed area and desk visibility. Continuous gaze-distribution feedback, a post-session summary, CSV logging, and timer integration complement the guidance system.

Talk MA Talk: Synthetic digital surface model generation from digital terrain models using stable diffusion

27.08.2026 14:00
G30

Speaker(s): Debwashis Borman

Digital surface models record buildings and vegetation on the bare earth. They are usually produced from airborne LiDAR, which is expensive to acquire, while terrain models and building footprints are widely available. Prior work has concentrated on the reverse problem, recovering the terrain from a surface model. To the author’s knowledge, the forward direction, generating the surface from terrain and footprints, has not previously been posed as a learned generative task. This thesis adapts Stable Diffusion 1.5 to metric elevation data. Its autoencoder is fine-tuned on elevation rasters, and two conditioning routes are compared. The first fine-tunes the full denoiser with the encoded terrain and footprint mask concatenated to the latent, and the second steers a frozen backbone through a ControlNet branch with low-rank adapters. A residual formulation that predicts only the height above the terrain and restores the real ground is evaluated alongside the absolute-surface target, and a from-scratch pix2pixHD GAN is the baseline. Nine models are trained on LiDAR data from Basel, Bern, and Zurich and judged twice, on realism with FID and KID, and on usefulness, by feeding the generated surfaces as training data to an independent surface-to-terrain model scored on a held-out region. Generation proves feasible, with plausible surfaces from every route and a best FID of 25.5. The central finding is that the two judgements invert. The most realistic model yields the least useful surfaces, the terrain-faithful residual and ControlNet routes the most useful, an ordering that repeats across both downstream regimes. No synthetic mixture beats real-only training on the held-out region, so the surfaces are ranked by their usefulness rather than credited with a net augmentation gain. A realism score alone therefore cannot certify a generated elevation surface, and a task-based evaluation belongs in the protocol whenever the output is a measurement rather than a picture.

Talk MA-Talk: Animating Fluids in Gaussian Splatting Scenes with 3D Motion Fields

24.07.2026 13:00
G30

Speaker(s): Kjell Keune

Talk MA-Talk: Saliency-Guided Assessing of Novel View Synthesis

29.06.2026 13:00
IZ G30 (Seminar Room)

Speaker(s): Carlotta Harms

Novel view synthesis (NVS) methods are typically evaluated with image-based quality metrics such as PSNR, SSIM, and LPIPS. However, these metrics do not always reflect how humans perceive visual quality, since artifacts are more relevant when they appear in visually attended image regions.

This thesis investigates whether visual saliency can improve the perceptual evaluation and optimization of NVS renderings. Using eye-tracking data and predicted saliency maps, saliency-guided variants of common quality metrics are compared against human preference judgments. The results show that saliency can improve the agreement between objective metrics and subjective perception, although the effect depends on the chosen metric and saliency mask. The thesis further explores saliency-guided training, showing how saliency-weighted losses can shift reconstruction quality toward perceptually relevant regions while introducing trade-offs in model complexity.

Talk Investigating Realism and Artifacts of AI Media Generation and Editing

29.04.2026 11:00 - 29.04.2026 12:00
G30 - Seminar room

Speaker(s): Leslie Wöhler

Generative AI could become a valuable tool for the creative industry, however, its application is met with several challenges. For creators, especially the lack of direct control of the output and the occurrence of artifacts limit the potential of the technology. Therefore, we set out to investigate the perception of generative AI for different types of media and study workflows that allow creatives to restrict the output of AI systems.

In this talk, we analyze how viewers notice AI, perceive artifacts, and interpret the output. We first look into the possibilities of generative AI-based editing for 360° photos and investigate how different display modalities affect viewers. Afterwards, we discuss the impact of motion and appearance artifacts for video generations and examine possible creation workflows.

Talk MA-Talk: Multi-View Coherent Rendering of Hierarchical 4D Gaussian-Based Representations for Non-Rigid Reconstruction

12.12.2025 13:00
G30

Speaker(s): Paula Wespe

Talk Bridging Graphics and Vision: Visual Computing Research at the Digital Future Lab

11.12.2025 13:00
G30

Speaker(s): Nick Michiels

Nick Michiels is a tenure-track Assistant Professor in the Faculty of Engineering Technology at Hasselt University. He is one of the leads of the Visual Computing research unit at the University's Digital Future Lab (DFL). The DFL is a core lab of Flanders Make, the strategic research centre for the manufacturing industry in Flanders.

In this talk, he will provide an overview of the main research initiatives of the Visual Computing group, highlighting how computer graphics and computer vision are increasingly converging. Through ongoing research projects and selected examples, he will illustrate how this synergy enables new methods and applications in visual computing, with a strong focus on industrial relevance.

Talk Promotion: Neural Reconstruction and Rendering of Dynamic Real-World Content from Monocular Video

10.10.2025 10:00 - 10.10.2025 12:00
IZ 161

Speaker(s): Moritz Kappel

Talk Neuronales Punkt-basiertes Rendering gestern und heute

10.10.2025 09:00 - 10.10.2025 09:45
G30

Speaker(s): Marc Stamminger

Punkt-basiertes Rendering wurde bereits vor 25 Jahren intensiv erforscht.

Mit dem Aufkommen von neuronalen Rendering-Techniken gewann das Thema vor kurzer Zeit jedoch wieder sehr an Bedeutung.

Vor allem mit „3D Gaussian Splatting“ gewannen viele alte Techniken wieder große Bedeutung.

In dem Vortrag wird diese Entwicklung dargestellt, und es werden Ergebnisse neuester Arbeiten auf diesem Gebiet gezeigt.

Talk MA-Talk: Moment-based Adaptive Fragment Density Reconstruction for Layered Order-Independent Transparency

12.09.2025 13:00
G30

Speaker(s): Ke Zhao

Talk BA-Talk: Designing and Implementing an Immersive Learning Space to Enhance Attention of Students with ADHD

03.09.2025 13:00
G30

Speaker(s): Nazli Cesur

Talk BA-Talk: Real-Time Control of Lighting Effects via Neural-Network-Based Detection of Salient Musical Features

01.08.2025 13:00
Hardstyle Lab

Speaker(s): Fakher Belkacem

This thesis describes the development of an interactive lighting control system that visualizes music in real-time. The system utilizes a Raspberry Pi, a microphone, and LEDs to process audio signals and generate visual effects. The core of the project is the implementation of a three-layer Deep Neural Network (DNN) trained directly on raw audio data in the time domain. This innovative approach differs from traditional methods that transform audio signals into the frequency domain.

The primary objective of the project was to apply theoretical knowledge in machine learning to a practical, hands-on project while overcoming unexpected challenges. Throughout the development process, various issues such as inconsistencies in data labeling and difficulties in model training were identified and resolved. These experiences highlighted the importance of maintaining a clear and simple vision during model training.

A notable achievement was the successful deployment of the model on a Raspberry Pi, demonstrating the system's capability to perform complex machine learning tasks on low-cost hardware. This work contributes to the fields of music visualization and interactive installations, offering potential applications in live performances and art exhibitions.

Talk Expanding Capabilities of 3D representations for XR

20.06.2025 11:00
G30

Speaker(s): Brent Zoomers

Novel view synthesis has seen rapid advancements, enabling the photorealistic rendering of unseen viewpoints from sparse input data. Among the latest methods, 3D Gaussian Splatting has emerged as a prominent technique due to its combination of high-fidelity reconstruction, fast training times, and editability thanks to its explicit representation. However, despite its academic success, real-world adoption remains limited. In this talk, Brent Zoomers explores the challenges that hinder the industrial application of such methods and presents his ongoing efforts to bridge this gap. He will share insights from his current research, discuss practical obstacles faced in real-world scenarios, and outline promising directions for future work aimed at making novel view synthesis viable for production settings.

Talk BA-Talk: Designing and Implementing an Augmented Reality Memory Palace to Enhance the Memory Performance of Individuals with ADHD

03.06.2025 15:00
G30

Speaker(s): Lara Maschkowitz

Children and adolescents with Attention Deficit Hyperactivity Disorder (ADHD) experience challenges in the academic domain, particularly in learning. This Bachelor’s thesis aims to prototypically implement a memory palace using Augmented Reality, tailored specifically to meet the needs of learners with ADHD. The target is to improve their memory performance. To achieve this, the concept is based on the existing literature and is implemented as a prototype, followed by a critical discussion of the concept and prototype.

Talk BA-Talk: Designing, Developing and Exploring Virtual Reality Mindfulness Interventions to Reduce Mind Wandering in Individuals with ADHD

03.06.2025 14:00
G30

Speaker(s): Hannes Ast

In dieser Bachelorarbeit werden Achtsamkeitsinterventionen in der virtuellen Realität (VR) entworfen, entwickelt und erforscht, die darauf abzielen, das Gedankenwandern bei Personen mit Aufmerksamkeitsdefizit-Hyperaktivitätsstörung (ADHS) zu verringern. Aufgrund von Kernsymptomen wie Unaufmerksamkeit, Impulsivität und Hyperaktivität stehen Menschen mit ADHS in akademischen Kontexten oft vor größeren Herausforderungen als Studierende ohne ADHS und profitieren von maßgeschneiderten Hilfsmitteln. Eine VR-Anwendung wurde entwickelt, um diese Probleme anzugehen, indem eine immersive und kontrollierte Umgebung geschaffen wurde, die externe Ablenkungen minimiert und dem Teilnehmer eine Aufgabe stellt, bei welcher sich auf eine bestimmte Sache konzentriert werden muss. Die Ergebnisse deuten darauf hin, dass VR-Achtsamkeit für einige Personen mit ADHS eine wirksame Methode sein kann, und unterstreichen die Bedeutung der Personalisierung.

Talk MA-Talk: Semantic Segmentation of Harz Dead Trees using Multi-temporal High Resolution Optical Imagery

21.05.2025 14:00
G30

Speaker(s): Aditya Murti

Forests play a crucial role in maintaining ecological balance, supporting biodiversity, and providing resources for human use. Monitoring forest health, particularly in the face of threats such as tree mortality, is essential for effective forest management and urban planning. Land use and land cover (LULC) maps help monitor the health of forests, including tracking deforestation, reforestation, and changes due to natural disturbances such as wildfires or bark beetle infestations. Remote sensing (RS) technology has emerged as a powerful tool for environmental monitoring, offering time-series data that can capture changes in forest conditions over time.

The primary objective of this research is to develop a novel DL model which includes the benefits of the aforementioned models for dead tree segmentation of multi-temporal remote sensing images of the Harz region. The DL model will then be applied to generate the segmentation maps of images of Harz region per month during the growing seasons of a year.

Talk Über Künstliche Intelligenz und Natürliche Dummheit

23.04.2025 18:30
Roter Saal im Schloss, Braunschweig

Speaker(s): Marcus Magnor

Akademie-Vorlesung im Schloss, Braunschweigische Wissenschaftliche Gesellschaft

Talk MA-Talk: Improving Hybrid-Transparency 3D Gaussian Splatting through Exact Volumetric Rendering of Ellipsoidal Primitives

23.04.2025 12:00
G30

Speaker(s): Mathias Ivanov

Talk MA-Talk: Transferring traditional learning approaches to an immersive VR environment to enhance executive functions in students with ADHD

20.03.2025 12:00 - 20.03.2025 13:00
IZ G30

Speaker(s): Florian Krüger

Talk MA-Talk: Perception-aware Color Reduction

20.03.2025 11:00
G30

Speaker(s): Jing Wang

Color reduction is an image processing technique designed to reduce memory consumption and save transmission resources. Traditional color reduction methods, such as k-means, may ignore the perceptual characteristics of the human visual system and result in poor image quality.

To improve the perceptual quality of the image, a perceptual loss based method is used to generate the color reduced image. This is more similar to the original image by optimising the color palettes and mapping strategies.

The aim of this thesis is to develop colour reduction methods that take into account the human visual system and human perception through perceptual loss functions, such as the Learned Perceptual Image Patch Similarity (LPIPS), to evaluate and optimise the color palette and mapping strategies.

Talk BA-Talk: Designing and Implementing a Serious Game for Teaching Concepts of Coding to Individuals with ADHD

31.01.2025 15:00 - 31.01.2025 16:00
IZ G30

Speaker(s): Joel Schaub

Talk Promotion: Ego-Motion Aware Immersive Rendering from Real-World Recorded Panorama Videos

31.01.2025 10:00 - 31.01.2025 12:00
IZ 161

Speaker(s): Moritz Mühlhausen

In this talk, we will explore how we can enhance the immersive experience in virtual reality (VR) by integrating natural motion effects — specifically, ego-motion-aware parallax effects — into real-world panoramic videos.

Traditional panoramic video allows users to view a scene in all directions, but it still limits the sense of presence, especially in VR, where true immersion requires not just looking around but also feeling as though you're moving within the space. This is where parallax comes in: the natural shift in perspective that occurs when we move our heads, which adds depth and realism to our surroundings.

The first part of this talk will focus on how we can use multiple panoramic images to simulate this motion effect. By applying image-warping techniques, we can approximate parallax, making the VR experience feel more dynamic and lifelike. Although this method doesn’t fully replicate the real-world motion, it significantly improves immersion.

Secondly, we will introduce a simpler but powerful approach using a single recording from a single stationary omnidirectional stereo (ODS) camera. This camera captures images for both the left and right eye simultaneously, providing built-in depth perception without the need for multiple cameras. This not only simplifies the capturing process but also allows for a more efficient creation of immersive VR content.

This talk will demonstrate how these methods, whether using multiple cameras or a single ODS camera, can improve depth perception and realism in VR applications. These innovations can make VR experiences — from gaming to education — more engaging and lifelike by offering an experience that feels more connected to real-world motion.

Talk MA-Talk: 4D Diffusion Priors for Robust Dynamic View Synthesis from Monocular Video

24.01.2025 13:00 - 24.01.2025 14:00
IZ G30

Speaker(s): Timon Scholz

Reconstructing dynamic scenes from only a single monocular input video for novel view synthesis is a severely under-constrained problem due to significant ambiguities in the inputs. Nonetheless, recent advancements in neural rendering have led to significant improvements in the field, especially when combined with high-quality priors for regularization. A recent prior that has been explored for static scene reconstruction is diffusion models, which are capable of generating photorealistic images even with very little guidance. Taming these models to provide multi-view-consistent outputs has proven to be challenging, but recent research has been able to generate promising results. Due to these challenges, adapting diffusion-based priors for dynamic novel view synthesis has been severely under-explored. For this reason, I adapt the recent ViewCrafter model for static scene reconstruction as a diffusion prior for the state-of-the-art dynamic neural rendering model D-NPC. The presented approach models each timestep of the input video separately, producing diffusion-based novel views for the entire sequence. I then utilize these views to regularize training gradients of the D-NPC model. By evaluating this approach both qualitatively and quantitatively, I am able to showcase promising results for improving the quality of the reconstructed scene. However, my findings also indicate that currently available diffusion models do not provide sufficient consistency among views to always provide benefits to the reconstruction. In fact, ViewCrafter fails to produce plausible results on many scenes and produces geometrically inconsistent results in many cases. This leads to excessive blurring in a lot of cases, oftentimes significantly decreasing visual quality. By documenting these challenges I hope to provide a baseline for future work to improve diffusion models for the use as scene reconstruction priors.