AI Paper Atlas

Research map / Point clouds / Shape and surface completion

DiffCD: A Symmetric Differentiable Chamfer Distance for Neural Implicit Surface Fitting

Linus Härenstam-Nielsen, Lu Sang, Abhishek Saroha, Nikita Araslanov, Daniel Cremers

ECCV 2024 · Point clouds › Shape and surface completion · approach: Distance-function surface reconstruction · 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 surface-to-points Chamfer term (symmetric DiffCD loss) to the IGR points-to-surface lossIGR · FAMOUS No NoiseCD squared: 3.092 -> 0.542Improved
use DiffCD instead of IGR loss under medium noiseIGR · FAMOUS Medium NoiseChamfer distance (CD): 1.454 -> 0.690Improved
use DiffCD instead of SIREN off-surface regularizer (mu=0.33)SIREN · Thingi10k No NoiseCD squared: 0.894 -> 0.580Improved
use DiffCD instead of SIREN (mu=0.33) on Chamfer angleSIREN · Thingi10k No NoiseChamfer angle (deg): 12.7 -> 13.0Hurt
increase SIREN SSA weight mu from 0.033 to 0.33 (ablation)SIREN · FAMOUS Medium NoiseChamfer distance (CD): 0.963 -> 0.739Improved
increase SIREN SSA weight mu from 0.033 to 0.33 on noise-free data (ablation)SIREN · FAMOUS No NoiseCD squared: 1.151 -> 1.525Hurt
increase SIREN weight mu (surface area regularization)SIREN · Dragon and Serapis shapesreconstruction errorMixed
compare DiffCD to supervised methods in medium noiseDiffCD · FAMOUS Medium NoiseChamfer distance (CD): NKSR 0.610 vs DiffCD 0.690Hurt
increase eikonal weight lambda to 0.5 and 1 for higher noise settingsDiffCD · FAMOUS Medium/Max Noisesurface qualityImproved
use DiffCD instead of IGR in compute costDiffCD · runtimetime per iteration: roughly 6x longerHurt

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