AI Paper Atlas

Research map / 3D vision / Face and shape

Differentiable Shadow Mapping for Efficient Inverse Graphics

Markus Worchel, Marc Alexa

CVPR 2023 · 3D vision › Face and shape · approach: Textured mesh generation · 8 citations (OpenAlex)

Read the paperCheck 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 changedWhereMeasuredEffect
Add differentiable shadow mapping to rasterizationRasterizer pose estimation · Monocular pose estimation, Dragon, 512 × 512rotation error (°): 8.02 -> 3.32Improved
Add differentiable shadow mapping to rasterizationRasterizer pose estimation · Monocular pose estimation, Spot, 512 × 512rotation error (°): 5.80 -> 0.05Improved
Use rasterization with shadow maps instead of path tracingMonocular pose estimation · Monocular pose estimation, Spot, 512 × 512runtime [s]: 381.80 -> 3.76Improved
Remove shadows and rely only on surface shading (ablation)Differentiable shadow mapping pose estimation · Monocular pose estimationpose accuracyHurt
Disable antialias smoothing on the shadow map and squared depth map (ablation)Differentiable shadow mapping · Single-occluder shadow reconstructionconvergence to target configurationHurt
Apply antialias smoothing to the shadow map and squared depth map (ablation)Differentiable shadow mapping · Single-occluder shadow reconstructionconvergence to target configurationImproved
Increase the filter kernel size (ablation)Differentiable shadow mapping · Single-occluder optimizationgradient-descent robustnessImproved
Increase the filter kernel size (ablation)Differentiable shadow mapping · Light direction estimationalignment accuracyNo clear effect
Increase the filter kernel size (ablation)Differentiable shadow mapping · Light direction estimationruntimeHurt
Increase shadow map resolution (ablation)Differentiable shadow mapping · Light direction estimationruntimeHurt
Use three shadows instead of one (ablation)Differentiable shadow mapping face reconstruction · Synthetic face reconstruction, mean across facesD_mean: 0.013 -> 0.008Improved
Use three shadows instead of one (ablation)Differentiable shadow mapping face reconstruction · Synthetic face reconstruction, mean across facesD_Hausdorff: 0.105 -> 0.068Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy