Show images of current Projects | Years: 2023 - 2024 - 2025.

Instant Visualization and Interaction for Large Point Clouds

Point clouds are a quintessential 3D geometry representation format, and often the first model obtained from reconstructive efforts, such as LIDAR scans. IVILPC aims for fast, authentic, interactive, and high-quality processing of such point-based data sets. Our project explores high-performance software rendering routines for various point-based primitives, such as point sprites, gaussian splats, surfels, and particle systems. Beyond conventional use cases, point cloud rendering also forms a key component of point-based machine learning methods and novel-view synthesis, where performance is paramount. We will exploit the flexibility and processing power of cutting-edge GPU architecture features to formulate novel, high-performance rendering approaches. The envisioned solutions will be applicable to unstructured point clouds for instant rendering of billions of points. Our research targets minimally-invasive compression, culling methods, and level-of-detail techniques for point-based rendering to deliver high performance and quality on-demand. We explore GPU-accelerated editing of point clouds, as well as common display issues on next-generation display devices. IVILPC lays the foundation for interaction with large point clouds in conventional and immersive environments. Its goal is an efficient data knowledge transfer from sensor to user, with a wide range of use cases to image-based rendering, virtual reality (VR) technology, architecture, the geospatial industry, and cultural heritage.

Modeling the World at Scale

Vision: reconstruct a model of the world that permits online level-of-detail extraction.

Photogrammetry made easy

Superhumans - Walking Through Walls

In recent years, virtual and augmented reality have gained widespread attention because of newly developed head-mounted displays. For the first time, mass-market penetration seems plausible. Also, range sensors are on the verge of being integrated into smartphones, evidenced by prototypes such as the Google Tango device, making ubiquitous on-line acquisition of 3D data a possibility. The combination of these two technologies – displays and sensors – promises applications where users can directly be immersed into an experience of 3D data that was just captured live. However, the captured data needs to be processed and structured before being displayed. For example, sensor noise needs to be removed, normals need to be estimated for local surface reconstruction, etc. The challenge is that these operations involve a large amount of data, and in order to ensure a lag-free user experience, they need to be performed in real time, i.e., in just a few milliseconds per frame. In this proposal, we exploit the fact that dynamic point clouds captured in real time are often only relevant for display and interaction in the current frame and inside the current view frustum. In particular, we propose a new view-dependent data structure that permits efficient connectivity creation and traversal of unstructured data, which will speed up surface recovery, e.g. for collision detection. Classifying occlusions comes at no extra cost, which will allow quick access to occluded layers in the current view. This enables new methods to explore and manipulate dynamic 3D scenes, overcoming interaction methods that rely on physics-based metaphors like walking or flying, lifting interaction with 3D environments to a “superhuman” level.

Real-Time Shape Acquisition with Sensor-Specific Precision

Acquiring shapes of physical objects in real time and with guaranteed precision to the noise model of the sensor devices.