AI Paper Atlas

Research map / 3D vision / Face and shape

Approximate Differentiable Rendering with Algebraic Surfaces

Leonid Keselman, Martial Hebert

ECCV 2022 · 3D vision › Face and shape · approach: Textured mesh generation · 37 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
use Fuzzy Metaballs renderer instead of SoftRas mesh rendererpose estimation · Pose estimation (noisy depth)pose error: 17.0 -> 4.2Improved
use Fuzzy Metaballs instead of high-fidelity point ICPpose estimation · Pose estimation (noisy depth)pose error: 6.8 -> 4.2Improved
use Fuzzy Metaballs instead of equal-fidelity point ICPpose estimation · Pose estimation (noise-free)pose error: 7.6 -> 4.0Improved
use Fuzzy Metaballs instead of Pulsarpose estimation · Pose estimation (noise-free)pose error: 20.2 -> 4.0Improved
use Fuzzy Metaballs instead of NeRFshape from silhouette · Shape from silhouette (noisy)silhouette cross-entropy: 0.062 -> 0.055Improved
use NeRF instead of Fuzzy Metaballsshape from silhouette · Shape from silhouette (noise-free)silhouette cross-entropy: 0.040 -> 0.032Hurt
use Fuzzy Metaballs instead of PyTorch3D meshshape from silhouette · Shape from silhouette (runtime)time (s): 3008 -> 68Improved
use Fuzzy Metaballs instead of voxel carving under noiseshape from silhouette · Shape from silhouette (noisy)silhouette cross-entropy: 1.119 ± 0.367 -> 0.055Improved
use Fuzzy Metaballs instead of SoftRas on CPUrendering · Rendering runtime (CPU, backward)ms: 384.7 -> 13.2Improved
lower alpha threshold from 50% to 10% (ablation)Fuzzy Metaballs · CO3D / phone video reconstructionqualitative recoveryImproved
tune PyTorch3D hyper-parameters with CMA-ES (ablation)PyTorch3D SoftRas baseline · pose estimation single framepose error: 16 degrees of rotation error and 9 percent of translation errorNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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