9 results
Search Results
Now showing 1 - 9 of 9
Item Dense 3D Gaussian Splatting Initialization for Sparse Image Data(The Eurographics Association, 2024) Seibt, Simon; Chang, Thomas Vincent Siu-Lung; von Rymon Lipinski, Bartosz ; Latoschik, Marc Erich; Liu, Lingjie; Averkiou, MelinosThis paper presents advancements in novel-view synthesis with 3D Gaussian Splatting (3DGS) using a dense and accurate SfM point cloud initialization approach. We address the challenge of achieving photorealistic renderings from sparse image data, where basic 3DGS training may result in suboptimal convergence, thus leading to visual artifacts. The proposed method enhances precision and density of initially reconstructed point clouds by refining 3D positions and extrapolating additional points, even for difficult image regions, e.g. with repeating patterns and suboptimal visual coverage. Our contributions focus on improving ''Dense Feature Matching for Structure-from-Motion'' (DFM4SfM) based on a homographic decomposition of the image space to support 3DGS training: First, a grid-based feature detection method is introduced for DFM4SfM to ensure a welldistributed 3D Gaussian initialization uniformly over all depth planes. Second, the SfM feature matching is complemented by a geometric plausibility check, priming the homography estimation and thereby improving the initial placement of 3D Gaussians. Experimental results on the NeRF-LLFF dataset demonstrate that this approach achieves superior qualitative and quantitative results, even for fewer views, and the potential for a significantly accelerated 3DGS training with faster convergence.Item Distributed Surface Reconstruction(The Eurographics Association, 2024) Marin, Diana; Komon, Patrick; Ohrhallinger, Stefan; Wimmer, Michael; Liu, Lingjie; Averkiou, MelinosRecent advancements in scanning technologies and their rise in availability have shifted the focus from reconstructing surfaces from point clouds of small areas to large, e.g., city-wide scenes, containing massive amounts of data. We adapt a surface reconstruction method to work in a distributed fashion on a high-performance cluster, reconstructing datasets with millions of vertices in seconds. We exploit the locality of the connectivity required by the reconstruction algorithm to efficiently divide-andconquer the problem of creating triangulations from very large unstructured point clouds.Item Recent Trends in 3D Reconstruction of General Non-Rigid Scenes(The Eurographics Association and John Wiley & Sons Ltd., 2024) Yunus, Raza; Lenssen, Jan Eric; Niemeyer, Michael; Liao, Yiyi; Rupprecht, Christian; Theobalt, Christian; Pons-Moll, Gerard; Huang, Jia-Bin; Golyanik, Vladislav; Ilg, Eddy; Aristidou, Andreas; Macdonnell, RachelReconstructing models of the real world, including 3D geometry, appearance, and motion of real scenes, is essential for computer graphics and computer vision. It enables the synthesizing of photorealistic novel views, useful for the movie industry and AR/VR applications. It also facilitates the content creation necessary in computer games and AR/VR by avoiding laborious manual design processes. Further, such models are fundamental for intelligent computing systems that need to interpret real-world scenes and actions to act and interact safely with the human world. Notably, the world surrounding us is dynamic, and reconstructing models of dynamic, non-rigidly moving scenes is a severely underconstrained and challenging problem. This state-of-the-art report (STAR) offers the reader a comprehensive summary of state-of-the-art techniques with monocular and multi-view inputs such as data from RGB and RGB-D sensors, among others, conveying an understanding of different approaches, their potential applications, and promising further research directions. The report covers 3D reconstruction of general non-rigid scenes and further addresses the techniques for scene decomposition, editing and controlling, and generalizable and generative modeling. More specifically, we first review the common and fundamental concepts necessary to understand and navigate the field and then discuss the state-of-the-art techniques by reviewing recent approaches that use traditional and machine-learning-based neural representations, including a discussion on the newly enabled applications. The STAR is concluded with a discussion of the remaining limitations and open challenges.Item D-NPC: Dynamic Neural Point Clouds for Non-Rigid View Synthesis from Monocular Video(The Eurographics Association and John Wiley & Sons Ltd., 2025) Kappel, Moritz; Hahlbohm, Florian; Scholz, Timon; Castillo, Susana; Theobalt, Christian; Eisemann, Martin; Golyanik, Vladislav; Magnor, Marcus; Bousseau, Adrien; Day, AngelaDynamic reconstruction and spatiotemporal novel-view synthesis of non-rigidly deforming scenes recently gained increased attention. While existing work achieves impressive quality and performance on multi-view or teleporting camera setups, most methods fail to efficiently and faithfully recover motion and appearance from casual monocular captures. This paper contributes to the field by introducing a new method for dynamic novel view synthesis from monocular video, such as casual smartphone captures. Our approach represents the scene as a dynamic neural point cloud, an implicit time-conditioned point distribution that encodes local geometry and appearance in separate hash-encoded neural feature grids for static and dynamic regions. By sampling a discrete point cloud from our model, we can efficiently render high-quality novel views using a fast differentiable rasterizer and neural rendering network. Similar to recent work, we leverage advances in neural scene analysis by incorporating data-driven priors like monocular depth estimation and object segmentation to resolve motion and depth ambiguities originating from the monocular captures. In addition to guiding the optimization process, we show that these priors can be exploited to explicitly initialize our scene representation to drastically improve optimization speed and final image quality. As evidenced by our experimental evaluation, our dynamic point cloud model not only enables fast optimization and real-time frame rates for interactive applications, but also achieves competitive image quality on monocular benchmark sequences. Our code and data are available online https://moritzkappel.github.io/projects/dnpc/.Item Efficient Perspective-Correct 3D Gaussian Splatting Using Hybrid Transparency(The Eurographics Association and John Wiley & Sons Ltd., 2025) Hahlbohm, Florian; Friederichs, Fabian; Weyrich, Tim; Franke, Linus; Kappel, Moritz; Castillo, Susana; Stamminger, Marc; Eisemann, Martin; Magnor, Marcus; Bousseau, Adrien; Day, Angela3D Gaussian Splats (3DGS) have proven a versatile rendering primitive, both for inverse rendering as well as real-time exploration of scenes. In these applications, coherence across camera frames and multiple views is crucial, be it for robust convergence of a scene reconstruction or for artifact-free fly-throughs. Recent work started mitigating artifacts that break multi-view coherence, including popping artifacts due to inconsistent transparency sorting and perspective-correct outlines of (2D) splats. At the same time, real-time requirements forced such implementations to accept compromises in how transparency of large assemblies of 3D Gaussians is resolved, in turn breaking coherence in other ways. In our work, we aim at achieving maximum coherence, by rendering fully perspective-correct 3D Gaussians while using a high-quality approximation of accurate blending, hybrid transparency, on a per-pixel level, in order to retain real-time frame rates. Our fast and perspectively accurate approach for evaluation of 3D Gaussians does not require matrix inversions, thereby ensuring numerical stability and eliminating the need for special handling of degenerate splats, and the hybrid transparency formulation for blending maintains similar quality as fully resolved per-pixel transparencies at a fraction of the rendering costs. We further show that each of these two components can be independently integrated into Gaussian splatting systems. In combination, they achieve up to 2× higher frame rates, 2× faster optimization, and equal or better image quality with fewer rendering artifacts compared to traditional 3DGS on common benchmarks.Item NoiseGS: Boosting 3D Gaussian Splatting with Positional Noise for Large-Scale Scene Rendering(The Eurographics Association, 2025) Kweon, Minseong; Cheng, Kai; Chen, Xuejin; Park, Jinsun; Ceylan, Duygu; Li, Tzu-Mao3D Gaussian Splatting (3DGS) efficiently renders 3D spaces by adaptively densifying anisotropic Gaussians from initial points. However, in complex scenes such as city-scale environments, large Gaussians often overlap with high-frequency regions rich in edges and fine details. In these areas, conflicting per-pixel gradient directions cause gradient cancellation, reducing the overall gradient magnitude and potentially causing Gaussians to remain trapped in suboptimal positions even after densification. To address this, we propose NoiseGS, a novel approach that integrates randomized noise injection into 3DGS, guiding suboptimal Gaussians selected for densification toward more optimal positions. In addition, to mitigate the instability caused by oversized Gaussians, we introduce an ℓp-penalization on the scale of Gaussians. Our method integrates seamlessly with existing heuristicbased optimization and demonstrates strong generalization in reconstructing complex scenes such as MatrixCity and Building.Item TemPCC: Completing Temporal Occlusions in Large Dynamic Point Clouds captured by Multiple RGB-D Cameras(The Eurographics Association, 2025) Mühlenbrock, Andre; Weller, Rene; Zachmann, Gabriel; Ceylan, Duygu; Li, Tzu-MaoWe present TemPCC, an approach to complete temporal occlusions in large dynamic point clouds. Our method manages a point set over time, integrates new observations into this set, and predicts the motion of occluded points based on the flow of surrounding visible ones. Unlike existing methods, our approach efficiently handles arbitrarily large point sets with linear complexity, does not reconstruct a canonical representation, and considers only local features. Our tests, performed on an Nvidia GeForce RTX 4090, demonstrate that our approach can complete a frame with 30,000 points in under 30 ms, while, in general, being able to handle point sets exceeding 1,000,000 points. This scalability enables the mitigation of temporal occlusions across entire scenes captured by multi-RGB-D camera setups. Our initial results demonstrate that self-occlusions are effectively completed and successfully generalized to unknown scenes despite limited training data.Item Learning Image Fractals Using Chaotic Differentiable Point Splatting(The Eurographics Association and John Wiley & Sons Ltd., 2025) Djeacoumar, Adarsh; Mujkanovic, Felix; Seidel, Hans-Peter; Leimkühler, Thomas; Bousseau, Adrien; Day, AngelaFractal geometry, defined by self-similar patterns across scales, is crucial for understanding natural structures. This work addresses the fractal inverse problem, which involves extracting fractal codes from images to explain these patterns and synthesize them at arbitrary finer scales. We introduce a novel algorithm that optimizes Iterated Function System parameters using a custom fractal generator combined with differentiable point splatting. By integrating both stochastic and gradient-based optimization techniques, our approach effectively navigates the complex energy landscapes typical of fractal inversion, ensuring robust performance and the ability to escape local minima. We demonstrate the method's effectiveness through comparisons with various fractal inversion techniques, highlighting its ability to recover high-quality fractal codes and perform extensive zoom-ins to reveal intricate patterns from just a single image.Item Learning Fast 3D Gaussian Splatting Rendering using Continuous Level of Detail(The Eurographics Association and John Wiley & Sons Ltd., 2025) Milef, Nicholas; Seyb, Dario; Keeler, Todd; Nguyen-Phuoc, Thu; Bozic, Aljaz; Kondguli, Sushant; Marshall, Carl; Bousseau, Adrien; Day, Angela3D Gaussian splatting (3DGS) has shown potential for rendering photorealistic 3D scenes in real-time. Unfortunately, rendering these scenes on less powerful hardware is still a challenge, especially with high-resolution displays. We introduce a continuous level of detail (CLOD) algorithm and demonstrate how our method can improve performance while preserving as much quality as possible. Our approach learns to order splats based on importance and optimize them such that a representative and realistic scene can be rendered for an arbitrary splat count. Our method does not require any additional memory or rendering overhead and works with existing 3DGS renderers. We also demonstrate the flexibility of our CLOD method by extending it with distance-based LOD selection, foveated rendering, and budget-based rendering.