Jeroen Put

Bram Vanherle, Jeroen Put, Nick Michiels et al.

In this paper a deep learning architecture is presented that can, in real time, detect the 2D locations of certain landmarks of physical tools, such as a hammer or screwdriver. To avoid the labor of manual labeling, the network is trained on synthetically generated data. Training computer vision models on computer generated images, while still achieving good accuracy on real images, is a challenge due to the difference in domain. The proposed method uses an advanced rendering method in combination with transfer learning and an intermediate supervision architecture to address this problem. It is shown that the model presented in this paper, named Intermediate Heatmap Model (IHM), generalizes to real images when trained on synthetic data. To avoid the need for an exact textured 3D model of the tool in question, it is shown that the model will generalize to an unseen tool when trained on a set of different 3D models of the same type of tool. IHM is compared to two existing approaches to keypoint detection and it is shown that it outperforms those at detecting tool landmarks, trained on synthetic data.

Brent Zoomers, Maarten Wijnants, Ivan Molenaers et al.

Over the past year, 3D Gaussian Splatting (3DGS) has received significant attention for its ability to represent 3D scenes in a perceptually accurate manner. However, it can require a substantial amount of storage since each splat's individual data must be stored. While compression techniques offer a potential solution by reducing the memory footprint, they still necessitate retrieving the entire scene before any part of it can be rendered. In this work, we introduce a novel approach for progressively rendering such scenes, aiming to display visible content that closely approximates the final scene as early as possible without loading the entire scene into memory. This approach benefits both on-device rendering applications limited by memory constraints and streaming applications where minimal bandwidth usage is preferred. To achieve this, we approximate the contribution of each Gaussian to the final scene and construct an order of prioritization on their inclusion in the rendering process. Additionally, we demonstrate that our approach can be combined with existing compression methods to progressively render (and stream) 3DGS scenes, optimizing bandwidth usage by focusing on the most important splats within a scene. Overall, our work establishes a foundation for making remotely hosted 3DGS content more quickly accessible to end-users in over-the-top consumption scenarios, with our results showing significant improvements in quality across all metrics compared to existing methods.

Bram Vanherle, Brent Zoomers, Jeroen Put et al.

Generating synthetic images is a useful method for cheaply obtaining labeled data for training computer vision models. However, obtaining accurate 3D models of relevant objects is necessary, and the resulting images often have a gap in realism due to challenges in simulating lighting effects and camera artifacts. We propose using the novel view synthesis method called Gaussian Splatting to address these challenges. We have developed a synthetic data pipeline for generating high-quality context-aware instance segmentation training data for specific objects. This process is fully automated, requiring only a video of the target object. We train a Gaussian Splatting model of the target object and automatically extract the object from the video. Leveraging Gaussian Splatting, we then render the object on a random background image, and monocular depth estimation is employed to place the object in a believable pose. We introduce a novel dataset to validate our approach and show superior performance over other data generation approaches, such as Cut-and-Paste and Diffusion model-based generation.

Joni Vanherck, Steven Moonen, Brent Zoomers et al.

Accurate camera localization is crucial for robotics and Extended Reality (XR), enabling reliable navigation and alignment of virtual and real content. Existing visual methods often suffer from drift, scale ambiguity, and depend on fiducials or loop closure. This work introduces a real-time method for localizing a camera within a pre-captured, highly accurate colored LiDAR point cloud. By rendering synthetic views from this cloud, 2D-3D correspondences are established between live frames and the point cloud. A neural rendering technique narrows the domain gap between synthetic and real images, reducing occlusion and background artifacts to improve feature matching. The result is drift-free camera tracking with correct metric scale in the global LiDAR coordinate system. Two real-time variants are presented: Online Render and Match, and Prebuild and Localize. We demonstrate improved results on the ScanNet++ dataset and outperform existing SLAM pipelines.