Learning-based scene representations such as neural radiance fields or light field networks, that rely on fitting a scene model to image observations, commonly encounter challenges in the presence of inconsistencies within the images caused by occlusions, inaccurately estimated camera parameters or effects like lens flare. To address this challenge, we introduce RANdom RAy Consensus (RANRAC), an efficient approach to eliminate the effect of inconsistent data, thereby taking inspiration from classical RANSAC based outlier detection for model fitting. In contrast to the down-weighting of the effect of outliers based on robust loss formulations, our approach reliably detects and excludes inconsistent perspectives, resulting in clean images without floating artifacts. For this purpose, we formulate a fuzzy adaption of the RANSAC paradigm, enabling its application to large scale models. We interpret the minimal number of samples to determine the model parameters as a tunable hyperparameter, investigate the generation of hypotheses with data-driven models, and analyse the validation of hypotheses in noisy environments. We demonstrate the compatibility and potential of our solution for both photo-realistic robust multi-view reconstruction from real-world images based on neural radiance fields and for single-shot reconstruction based on light-field networks. In particular, the results indicate significant improvements compared to state-of-the-art robust methods for novel-view synthesis on both synthetic and captured scenes with various inconsistencies including occlusions, noisy camera pose estimates, and unfocused perspectives. The results further indicate significant improvements for single-shot reconstruction from occluded images.

We propose a novel cross-spectral rendering framework based on 3D Gaussian Splatting (3DGS) that generates realistic and semantically meaningful splats from registered multi-view spectrum and segmentation maps. This extension enhances the representation of scenes with multiple spectra, providing insights into the underlying materials and segmentation. We introduce an improved physically-based rendering approach for Gaussian splats, estimating reflectance and lights per spectra, thereby enhancing accuracy and realism. In a comprehensive quantitative and qualitative evaluation, we demonstrate the superior performance of our approach with respect to other recent learning-based spectral scene representation approaches (i.e., XNeRF and SpectralNeRF) as well as other non-spectral state-of-the-art learning-based approaches. Our work also demonstrates the potential of spectral scene understanding for precise scene editing techniques like style transfer, inpainting, and removal. Thereby, our contributions address challenges in multi-spectral scene representation, rendering, and editing, offering new possibilities for diverse applications.

Spectral rendering accurately simulates light-material interactions by considering the entire light spectrum, unlike traditional rendering methods that use limited color channels like RGB. This technique is particularly valuable in industries to assess visual quality before production. Moreover, Spectral imaging finds extensive applications in fields like agriculture for plant disease detection, cultural heritage for preservation, forensic science, environment monitoring and medical science among others. Advances in generating novel views from images have been achieved through methods like NERF and Gaussian splatting, which outperforms others in terms of quality. This paper introduces a web-based viewer built on the Viser framework for visualizing and comparing cross-spectral Gaussian splats from different views and during various training stages. This viewer supports real-time collaboration and comprehensive visual comparison, enhancing user experience in spectral data analysis. We conduct a user study and performance analysis to confirm its effectiveness and usability for different application scenarios, while also proposing potential enhancements for increased functionality.

In this paper, we focus on investigating the potential of advanced Neural Radiance Fields (NeRFs) and 3D Gaussian Splatting for 3D scene reconstruction from aerial imagery obtained via sensor platforms with an almost nadir-looking camera. Such a setting for image acquisition is convenient for capturing large-scale urban scenes, yet it poses particular challenges arising from imagery with large overlap, very short baselines, similar viewing direction and almost the same but large distance to the scene, and it therefore differs from the usual object-centric scene capture. We apply a traditional approach for image-based 3D reconstruction (COLMAP), a modern NeRF-based approach (Nerfacto) and a representative for the recently introduced 3D Gaussian Splatting approaches (Splatfacto), where the latter two are provided in the Nerfstudio framework. We analyze results achieved on the recently released UseGeo dataset both quantitatively and qualitatively. The achieved results reveal that the traditional COLMAP approach still outperforms Nerfacto and Splatfacto approaches for various scene characteristics, such as less-textured areas, areas with high vegetation, shadowed areas and areas observed from only very few views.

We present incomplete gamma kernels, a generalization of Locally Optimal Projection (LOP) operators. In particular, we reveal the relation of the classical localized L1 estimator, used in the LOP operator for point cloud denoising, to the common Mean Shift framework via a novel kernel. Furthermore, we generalize this result to a whole family of kernels that are built upon the incomplete gamma function and each represents a localized Lp estimator. By deriving various properties of the kernel family concerning distributional, Mean Shift induced, and other aspects such as strict positive definiteness, we obtain a deeper understanding of the operator's projection behavior. From these theoretical insights, we illustrate several applications ranging from an improved Weighted LOP (WLOP) density weighting scheme and a more accurate Continuous LOP (CLOP) kernel approximation to the definition of a novel set of robust loss functions. These incomplete gamma losses include the Gaussian and LOP loss as special cases and can be applied to various tasks including normal filtering. Furthermore, we show that the novel kernels can be included as priors into neural networks. We demonstrate the effects of each application in a range of quantitative and qualitative experiments that highlight the benefits induced by our modifications.

Novel view synthesis (NVS) of urban scenes enables the exploration of cities virtually and interactively, which can further be used for urban planning, navigation, digital tourism, etc. However, many current NVS methods require a large amount of images from known views as input and are sensitive to intrinsic and extrinsic camera parameters. In this paper, we propose a new unified framework for NVS of urban scenes with fewer required views via the integration of scene priors and the joint optimization of camera parameters under an geometric constraint along with NeRF weights. The integration of scene priors makes full use of the priors from the neighbor reference views to reduce the number of required known views. The joint optimization can correct the errors in camera parameters, which are usually derived from algorithms like Structure-from-Motion (SfM), and then further improves the quality of the generated novel views. Experiments show that our method achieves about 25.375 dB and 25.512 dB in average in terms of peak signal-to-noise (PSNR) on synthetic and real data, respectively. It outperforms popular state-of-the-art methods (i.e., BungeeNeRF and MegaNeRF) by about 2–4 dB in PSNR. Notably, our method achieves better or competitive results than the baseline method with only one third of the known view images required for the baseline. The code and dataset are available at https://github.com/Dongber/PriNeRF.

Whereas emerging learning-based scene representations are predominantly evaluated based on image quality metrics such as PSNR, SSIM or LPIPS, only a few investigations focus on the evaluation of geometric accuracy of the underlying model. In contrast to only demonstrating the geometric deviations of models for the fully optimized scene model, our work aims at investigating the geometric convergence behavior during the optimization. For this purpose, we analyze the geometric convergence of discretized density fields by leveraging respectively derived point cloud representations for different training steps during the optimization of the scene representation and their comparison based on established point cloud metrics, thereby allowing insights regarding which scene parts are already represented well within the scene representation at a certain time during the optimization. By demonstrating that certain regions reach convergence earlier than other regions in the scene, we provide the motivation regarding future developments on locally-guided optimization approaches to shift the computational burden to the adjustment of regions that still need to converge while leaving converged regions unchanged which might help to further reduce training time and improve the achieved quality.

Accurately replicating the appearance of real-world materials in computer graphics is a complex task due to the intricate interactions between light, reflectance, and geometry. In this paper we address the challenges of material representation, acquisition, and editing by leveraging the potential of deep learning algorithms our framework provide. To enable the visualization and generation of material assets from single or multi-view images, allowing for the estimation of materials from real world objects. Additionally, a material asset exporter, enabling the export of materials in widely used formats and facilitating easy editing using common content creator tools. The proposed framework enables designers to effectively collaborate and seamlessly integrate deep learning-based material estimation models into their design pipelines using traditional content creation tools. An analysis of the performance and memory usage of material assets at various texture resolutions shows that our framework can be used plausibly according to the needs of the end-user.

We investigate the capabilities of neural inverse procedural modeling to infer high-quality procedural yarn models with fiber-level details from single images of depicted yarn samples. While directly inferring all parameters of the underlying yarn model based on a single neural network may seem an intuitive choice, we show that the complexity of yarn structures in terms of twisting and migration characteristics of the involved fibers can be better encountered in terms of ensembles of networks that focus on individual characteristics. We analyze the effect of different loss functions including a parameter loss to penalize the deviation of inferred parameters to ground truth annotations, a reconstruction loss to enforce similar statistics of the image generated for the estimated parameters in comparison to training images as well as an additional regularization term to explicitly penalize deviations between latent codes of synthetic images and the average latent code of real images in the encoder's latent space. We demonstrate that the combination of a carefully designed parametric, procedural yarn model with respective network ensembles as well as loss functions even allows robust parameter inference when solely trained on synthetic data. Since our approach relies on the availability of a yarn database with parameter annotations and we are not aware of such a respectively available dataset, we additionally provide, to the best of our knowledge, the first dataset of yarn images with annotations regarding the respective yarn parameters. For this purpose, we use a novel yarn generator that improves the realism of the produced results over previous approaches.

Extracting high-level structural information from 3D point clouds is challenging but essential for tasks like urban planning or autonomous driving requiring an advanced understanding of the scene at hand. Existing approaches are still not able to produce high-quality results consistently while being fast enough to be deployed in scenarios requiring interactivity. We propose to utilize a novel set of features describing the local neighborhood on a per-point basis via first and second order statistics as input for a simple and compact classification network to distinguish between non-edge, sharp-edge, and boundary points in the given data. Leveraging this feature embedding enables our algorithm to outperform the state-of-the-art technique PCEDNet in terms of quality and processing time while additionally allowing for the detection of boundaries in the processed point clouds.

We propose an efficient and GPU-accelerated sampling framework which enables unbiased gradient approximation for differentiable point cloud rendering based on surface splatting. Our framework models the contribution of a point to the rendered image as a probability distribution. We derive an unbiased approximative gradient for the rendering function within this model. To efficiently evaluate the proposed sample estimate, we introduce a tree-based data-structure which employs multipole methods to draw samples in near linear time. Our gradient estimator allows us to avoid regularization required by previous methods, leading to a more faithful shape recovery from images. Furthermore, we validate that these improvements are applicable to real-world applications by refining the camera poses and point cloud obtained from a real-time SLAM system. Finally, employing our framework in a neural rendering setting optimizes both the point cloud and network parameters, highlighting the framework’s ability to enhance data driven approaches.

The Microsoft HoloLens is a head-worn mobile augmented reality device. It allows a real-time 3D mapping of its direct environment and a self-localisation within the acquired 3D data. Both aspects are essential for robustly augmenting the local environment around the user with virtual contents and for the robust interaction of the user with virtual objects. Although not primarily designed as an indoor mapping device, the Microsoft HoloLens has a high potential for an efficient and comfortable mapping of both room-scale and building-scale indoor environments. In this paper, we provide a survey on the capabilities of the Microsoft HoloLens (Version 1) for the efficient 3D mapping and modelling of indoor scenes. More specifically, we focus on its capabilities regarding the localisation (in terms of pose estimation) within indoor environments and the spatial mapping of indoor environments. While the Microsoft HoloLens can certainly not compete in providing highly accurate 3D data like laser scanners, we demonstrate that the acquired data provides sufficient accuracy for a subsequent standard rule-based reconstruction of a semantically enriched and topologically correct model of an indoor scene from the acquired data. Furthermore, we provide a discussion with respect to the robustness of standard handcrafted geometric features extracted from data acquired with the Microsoft HoloLens and typically used for a subsequent learning-based semantic segmentation.