AI Paper Atlas

Research map / Point clouds / Shape and surface completion

SA-ConvONet: Sign-Agnostic Optimization of Convolutional Occupancy Networks

Jiapeng Tang, Jiabao Lei, Dan Xu, Feiying Ma, Kui Jia, Lei Zhang

ICCV 2021 · Point clouds › Shape and surface completion · approach: Distance-function surface reconstruction · 78 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
optimize the whole network instead of only the encoder (ablation)SA-ConvONet · ShapeNet-chairChamfer Distance: 0.516 -> 0.522No clear effect
optimize the whole network instead of only the encoder (ablation)SA-ConvONet · Synthetic RoomChamfer Distance: 0.516 -> 0.495Improved
optimize the whole network instead of only the encoder (ablation)SA-ConvONet · ScanNetChamfer Distance: 0.741 -> 0.728Improved
add convolutional occupancy pre-training instead of SAL geometric (sphere) initialization (ablation)SA-ConvONet · ShapeNet-chair (qualitative)reconstruction qualityImproved
reduce input from 30,000 to 5,000 points (ablation)SA-ConvONet · ShapeNet-chairChamfer Distance: 0.522 -> 0.529No clear effect
reduce input from 30,000 to 5,000 points (ablation)SA-ConvONet · Synthetic RoomChamfer Distance: 0.495 -> 0.511No clear effect
increase input to 50,000 points (ablation)SA-ConvONet · Synthetic RoomChamfer Distance: 0.495 -> 0.476No clear effect
sign-agnostic test-time optimization of the whole network vs CONet without optimizationSA-ConvONet · ShapeNet-chairChamfer Distance: 0.821 -> 0.522Improved
sign-agnostic test-time optimization vs CONetSA-ConvONet · Synthetic RoomChamfer Distance: 2.020 -> 0.495Improved
sign-agnostic test-time optimization vs CONetSA-ConvONet · ScanNetChamfer Distance: 1.559 -> 0.728Improved
sign-agnostic optimization vs SPSR (needs normals)SA-ConvONet · ScanNetF-score (tau): 82.33 -> 82.08Hurt

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