Research map / 3D vision / Face and shape
GenDR: A Generalized Differentiable Renderer
Felix Petersen, Bastian Goldluecke, Christian Borgelt, Oliver Deussen
Check your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use the uniform distribution with probabilistic aggregation | GenDR · ShapeNet 13-class single-view 3D reconstruction | mean 3D IoU: 0.6232 | Improved |
| Choose the smoothing distribution according to the task, class, and data characteristics rather than fixing one distribution | GenDR · ShapeNet 13-class single-view 3D reconstruction | 3D IoU | Mixed |
| Use regular distributions instead of their squared-distance counterparts | GenDR shape optimization · Airplane shape optimization | optimization performance | Improved |
| Use exponentially converging distributions rather than linearly converging ones | GenDR shape optimization · Airplane shape optimization | optimization performance | Improved |
| Use reversed asymmetric distributions rather than the original distributions when their mass otherwise shrinks faces | GenDR · Airplane shape optimization | optimization performance | Improved |
| Use extreme-parameter T-conorm variants such as Yager (p=4), Aczél-Alsina (p=2 or p=4), Dombi (p=2 or p=4), and Schweizer-Sklar (p=-2 or p=-4) | GenDR shape optimization · Airplane shape optimization | optimization performance | Hurt |
| Use Yager T-conorm with p=2 | GenDR shape optimization · Airplane shape optimization | optimization performance | Improved |
| Vary the training learning rate (ablation) | ShapeNet single-view reconstruction training · ShapeNet 13-class single-view 3D reconstruction | 3D reconstruction performance | No clear effect |
Closest papers by meaning
- A General Differentiable Mesh Renderer for Image-Based 3D Reasoning (TPAMI 2020 · similarity 0.91)
- Soft Rasterizer: A Differentiable Renderer for Image-Based 3D Reasoning (ICCV 2019 · similarity 0.90)
- Differentiable rendering with perturbed optimizers (NeurIPS 2021 · similarity 0.90)
- DRWR: A Differentiable Renderer without Rendering for Unsupervised 3D Structure Learning from Silhouette Images (ICML 2020 · similarity 0.88)
- LinPrim: Linear Primitives for Differentiable Volumetric Rendering (NeurIPS 2025 · similarity 0.86)
- BG-Triangle: Bezier Gaussian Triangle for 3D Vectorization and Rendering (CVPR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
