Real-Time Multi-Material Reflectance Reconstruction for Large-Scale Scenes Under Uncontrolled Illumination from RGB-D Image Sequences

Bode, Lukas; Merzbach, Sebastian; Stotko, Patrick; Weinmann, Michael; Klein, Reinhard

doi:10.1109/3dv.2019.00083

Cited by 1 publication

(1 citation statement)

References 142 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Application -Room-Scale Scene Refinement. To demonstrate that our meth-od scales well to room-scale scenes, we utilize sequence RGB-D datasets by Steinbrucker et al [59] and Bode et al [6]. The latter registers RGB and depth measurements by masking RGB pixels which have no corresponding depth values.…”

Section: Applications and Resultsmentioning

confidence: 99%

Unbiased Gradient Estimation for Differentiable Surface Splatting via Poisson Sampling

Müller

Weinmann

Klein

2022

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

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 realworld 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.

show abstract

Section: Applications and Resultsmentioning

confidence: 99%