Marcus Magnor

Moritz Kappel, Vladislav Golyanik, Susana Castillo et al.

The reconstruction and novel view synthesis of dynamic scenes recently gained increased attention. As reconstruction from large-scale multi-view data involves immense memory and computational requirements, recent benchmark datasets provide collections of single monocular views per timestamp sampled from multiple (virtual) cameras. We refer to this form of inputs as "monocularized" data. Existing work shows impressive results for synthetic setups and forward-facing real-world data, but is often limited in the training speed and angular range for generating novel views. This paper addresses these limitations and proposes a new method for full 360° inward-facing novel view synthesis of non-rigidly deforming scenes. At the core of our method are: 1) An efficient deformation module that decouples the processing of spatial and temporal information for accelerated training and inference; and 2) A static module representing the canonical scene as a fast hash-encoded neural radiance field. In addition to existing synthetic monocularized data, we systematically analyze the performance on real-world inward-facing scenes using a newly recorded challenging dataset sampled from a synchronized large-scale multi-view rig. In both cases, our method is significantly faster than previous methods, converging in less than 7 minutes and achieving real-time framerates at 1K resolution, while obtaining a higher visual accuracy for generated novel views. Our source code and data is available at our project page https://graphics.tu-bs.de/publications/kappel2022fast.

Felix Lehner, Pasquale Lombardo, Susana Castillo et al.

We present an in-depth analysis on how to build and train neural networks to estimate the spatial distribution of scattered radiation fields for radiation protection dosimetry in medical radiation fields, such as those found in interventional radiology and cardiology. We present three different synthetically generated datasets with increasing complexity for training, using a Monte-Carlo Simulation application based on Geant4. On those datasets, we evaluate convolutional and fully connected architectures of neural networks to demonstrate which design decisions work well for reconstructing the fluence and spectra distributions over the spatial domain of such radiation fields. All our datasets, as well as our training pipeline, are published as open source in separate repositories.

Florian Hahlbohm, Linus Franke, Martin Eisemann et al.

Recent advances in 3D Gaussian Splatting (3DGS) have focused on accelerating optimization while preserving reconstruction quality. However, many proposed methods entangle implementation-level improvements with fundamental algorithmic modifications or trade performance for fidelity, leading to a fragmented research landscape that complicates fair comparison. In this work, we consolidate and evaluate the most effective and broadly applicable strategies from prior 3DGS research and augment them with several novel optimizations. We further investigate underexplored aspects of the framework, including numerical stability, Gaussian truncation, and gradient approximation. The resulting system, Faster-GS, provides a rigorously optimized algorithm that we evaluate across a comprehensive suite of benchmarks. Our experiments demonstrate that Faster-GS achieves up to 5$\times$ faster training while maintaining visual quality, establishing a new cost-effective and resource efficient baseline for 3DGS optimization. Furthermore, we demonstrate that optimizations can be applied to 4D Gaussian reconstruction, leading to efficient non-rigid scene optimization.

Felix Lehner, Pasquale Lombardo, Susana Castillo et al.

In this research work, we present our open-source Geant4-based Monte-Carlo simulation application, called RadField3D, for generating threedimensional radiation field datasets for dosimetry. Accompanying, we introduce a fast, machine-interpretable data format with a Python API for easy integration into neural network research, that we call RadFiled3D. Both developments are intended to be used to research alternative radiation simulation methods using deep learning.

Florian Hahlbohm, Linus Franke, Moritz Kappel et al.

We introduce a new approach for reconstruction and novel view synthesis of unbounded real-world scenes. In contrast to previous methods using either volumetric fields, grid-based models, or discrete point cloud proxies, we propose a hybrid scene representation, which implicitly encodes the geometry in a continuous octree-based probability field and view-dependent appearance in a multi-resolution hash grid. This allows for extraction of arbitrary explicit point clouds, which can be rendered using rasterization. In doing so, we combine the benefits of both worlds and retain favorable behavior during optimization: Our novel implicit point cloud representation and differentiable bilinear rasterizer enable fast rendering while preserving the fine geometric detail captured by volumetric neural fields. Furthermore, this representation does not depend on priors like structure-from-motion point clouds. Our method achieves state-of-the-art image quality on common benchmarks. Furthermore, we achieve fast inference at interactive frame rates, and can convert our trained model into a large, explicit point cloud to further enhance performance.

Brent Zoomers, Florian Hahlbohm, Joni Vanherck et al.

3D Gaussian Splatting can exploit frustum culling and level-of-detail strategies to accelerate rendering of scenes containing a large number of primitives. However, the semi-transparent nature of Gaussians prevents the application of another highly effective technique: occlusion culling. We address this limitation by proposing a novel method to learn the viewpoint-dependent visibility function of all Gaussians in a trained model using a small, shared MLP across instances of an asset in a scene. By querying it for Gaussians within the viewing frustum prior to rasterization, our method can discard occluded primitives during rendering. Leveraging Tensor Cores for efficient computation, we integrate these neural queries directly into a novel instanced software rasterizer. Our approach outperforms the current state of the art for composed scenes in terms of VRAM usage and image quality, utilizing a combination of our instanced rasterizer and occlusion culling MLP, and exhibits complementary properties to existing LoD techniques.

Florian Hahlbohm, Linus Franke, Leon Overkämping et al.

Implicit Neural Point Cloud (INPC) is a recent hybrid representation that combines the expressiveness of neural fields with the efficiency of point-based rendering, achieving state-of-the-art image quality in novel view synthesis. However, as with other high-quality approaches that query neural networks during rendering, the practical usability of INPC is limited by comparatively slow rendering. In this work, we present a collection of optimizations that significantly improve both the training and inference performance of INPC without sacrificing visual fidelity. The most significant modifications are an improved rasterizer implementation, more effective sampling techniques, and the incorporation of pre-training for the convolutional neural network used for hole-filling. Furthermore, we demonstrate that points can be modeled as small Gaussians during inference to further improve quality in extrapolated, e.g., close-up views of the scene. We design our implementations to be broadly applicable beyond INPC and systematically evaluate each modification in a series of experiments. Our optimized INPC pipeline achieves up to 25% faster training, 2x faster rendering, and 20% reduced VRAM usage paired with slight image quality improvements.

Moritz Kappel, Florian Hahlbohm, Timon Scholz et al.

Dynamic 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 $\textit{dynamic neural point cloud}$, 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/.

Marc Kassubeck, Moritz Kappel, Susana Castillo et al.

This paper deals with the highly challenging problem of reconstructing the shape of a refracting object from a single image of its resulting caustic. Due to the ubiquity of transparent refracting objects in everyday life, reconstruction of their shape entails a multitude of practical applications. The recent Shape from Caustics (SfC) method casts the problem as the inverse of a light propagation simulation for synthesis of the caustic image, that can be solved by a differentiable renderer. However, the inherent complexity of light transport through refracting surfaces currently limits the practicability with respect to reconstruction speed and robustness. To address these issues, we introduce Neural-Shape from Caustics (N-SfC), a learning-based extension that incorporates two components into the reconstruction pipeline: a denoising module, which alleviates the computational cost of the light transport simulation, and an optimization process based on learned gradient descent, which enables better convergence using fewer iterations. Extensive experiments demonstrate the effectiveness of our neural extensions in the scenario of quality control in 3D glass printing, where we significantly outperform the current state-of-the-art in terms of computational speed and final surface error.

Felix Klose, Oliver Wang, Jean-Charles Bazin et al.

Many compelling video processing effects can be achieved if per-pixel depth information and 3D camera calibrations are known. However, the success of such methods is highly dependent on the accuracy of this "scene-space" information. We present a novel, sampling-based framework for processing video that enables high-quality scene-space video effects in the presence of inevitable errors in depth and camera pose estimation. Instead of trying to improve the explicit 3D scene representation, the key idea of our method is to exploit the high redundancy of approximate scene information that arises due to most scene points being visible multiple times across many frames of video. Based on this observation, we propose a novel pixel gathering and filtering approach. The gathering step is general and collects pixel samples in scene-space, while the filtering step is application-specific and computes a desired output video from the gathered sample sets. Our approach is easily parallelizable and has been implemented on GPU, allowing us to take full advantage of large volumes of video data and facilitating practical runtimes on HD video using a standard desktop computer. Our generic scene-space formulation is able to comprehensively describe a multitude of video processing applications such as denoising, deblurring, super resolution, object removal, computational shutter functions, and other scene-space camera effects. We present results for various casually captured, hand-held, moving, compressed, monocular videos depicting challenging scenes recorded in uncontrolled environments.

Moritz Kappel, Vladislav Golyanik, Mohamed Elgharib et al.

Video-based human motion transfer creates video animations of humans following a source motion. Current methods show remarkable results for tightly-clad subjects. However, the lack of temporally consistent handling of plausible clothing dynamics, including fine and high-frequency details, significantly limits the attainable visual quality. We address these limitations for the first time in the literature and present a new framework which performs high-fidelity and temporally-consistent human motion transfer with natural pose-dependent non-rigid deformations, for several types of loose garments. In contrast to the previous techniques, we perform image generation in three subsequent stages, synthesizing human shape, structure, and appearance. Given a monocular RGB video of an actor, we train a stack of recurrent deep neural networks that generate these intermediate representations from 2D poses and their temporal derivatives. Splitting the difficult motion transfer problem into subtasks that are aware of the temporal motion context helps us to synthesize results with plausible dynamics and pose-dependent detail. It also allows artistic control of results by manipulation of individual framework stages. In the experimental results, we significantly outperform the state-of-the-art in terms of video realism. Our code and data will be made publicly available.

Thiemo Alldieck, Gerard Pons-Moll, Christian Theobalt et al.

We present a simple yet effective method to infer detailed full human body shape from only a single photograph. Our model can infer full-body shape including face, hair, and clothing including wrinkles at interactive frame-rates. Results feature details even on parts that are occluded in the input image. Our main idea is to turn shape regression into an aligned image-to-image translation problem. The input to our method is a partial texture map of the visible region obtained from off-the-shelf methods. From a partial texture, we estimate detailed normal and vector displacement maps, which can be applied to a low-resolution smooth body model to add detail and clothing. Despite being trained purely with synthetic data, our model generalizes well to real-world photographs. Numerous results demonstrate the versatility and robustness of our method.

Thiemo Alldieck, Marcus Magnor, Bharat Lal Bhatnagar et al.

We present a learning-based model to infer the personalized 3D shape of people from a few frames (1-8) of a monocular video in which the person is moving, in less than 10 seconds with a reconstruction accuracy of 5mm. Our model learns to predict the parameters of a statistical body model and instance displacements that add clothing and hair to the shape. The model achieves fast and accurate predictions based on two key design choices. First, by predicting shape in a canonical T-pose space, the network learns to encode the images of the person into pose-invariant latent codes, where the information is fused. Second, based on the observation that feed-forward predictions are fast but do not always align with the input images, we predict using both, bottom-up and top-down streams (one per view) allowing information to flow in both directions. Learning relies only on synthetic 3D data. Once learned, the model can take a variable number of frames as input, and is able to reconstruct shapes even from a single image with an accuracy of 6mm. Results on 3 different datasets demonstrate the efficacy and accuracy of our approach.

Thiemo Alldieck, Marcus Magnor, Weipeng Xu et al.

We present a novel method for high detail-preserving human avatar creation from monocular video. A parameterized body model is refined and optimized to maximally resemble subjects from a video showing them from all sides. Our avatars feature a natural face, hairstyle, clothes with garment wrinkles, and high-resolution texture. Our paper contributes facial landmark and shading-based human body shape refinement, a semantic texture prior, and a novel texture stitching strategy, resulting in the most sophisticated-looking human avatars obtained from a single video to date. Numerous results show the robustness and versatility of our method. A user study illustrates its superiority over the state-of-the-art in terms of identity preservation, level of detail, realism, and overall user preference.

Thiemo Alldieck, Marcus Magnor, Weipeng Xu et al.

This paper describes how to obtain accurate 3D body models and texture of arbitrary people from a single, monocular video in which a person is moving. Based on a parametric body model, we present a robust processing pipeline achieving 3D model fits with 5mm accuracy also for clothed people. Our main contribution is a method to nonrigidly deform the silhouette cones corresponding to the dynamic human silhouettes, resulting in a visual hull in a common reference frame that enables surface reconstruction. This enables efficient estimation of a consensus 3D shape, texture and implanted animation skeleton based on a large number of frames. We present evaluation results for a number of test subjects and analyze overall performance. Requiring only a smartphone or webcam, our method enables everyone to create their own fully animatable digital double, e.g., for social VR applications or virtual try-on for online fashion shopping.

Thiemo Alldieck, Marc Kassubeck, Marcus Magnor

We present a generative method to estimate 3D human motion and body shape from monocular video. Under the assumption that starting from an initial pose optical flow constrains subsequent human motion, we exploit flow to find temporally coherent human poses of a motion sequence. We estimate human motion by minimizing the difference between computed flow fields and the output of an artificial flow renderer. A single initialization step is required to estimate motion over multiple frames. Several regularization functions enhance robustness over time. Our test scenarios demonstrate that optical flow effectively regularizes the under-constrained problem of human shape and motion estimation from monocular video.

Fred W. Rauskolb, Kai Berger, Christian Lipski et al.

The 2007 DARPA Urban Challenge afforded the golden opportunity for the Technische Universität Braunschweig to demonstrate its abilities to develop an autonomously driving vehicle to compete with the world's best competitors. After several stages of qualification, our team CarOLO qualified early for the DARPA Urban Challenge Final Event and was among only eleven teams from initially 89 competitors to compete in the final. We had the ability to work together in a large group of experts, each contributing his expertise in his discipline, and significant organisational, financial and technical support by local sponsors who helped us to become the best non-US team. In this report, we describe the 2007 DARPA Urban Challenge, our contribution "Caroline", the technology and algorithms along with her performance in the DARPA Urban Challenge Final Event on November 3, 2007.

Dirk A. Lorenz, Stephan Wenger, Frank Schöpfer et al.

An algorithmic framework to compute sparse or minimal-TV solutions of linear systems is proposed. The framework includes both the Kaczmarz method and the linearized Bregman method as special cases and also several new methods such as a sparse Kaczmarz solver. The algorithmic framework has a variety of applications and is especially useful for problems in which the linear measurements are slow and expensive to obtain. We present examples for online compressed sensing, TV tomographic reconstruction and radio interferometry.