AI Paper Atlas

Research map / Point clouds / Segmentation and detection

GaPro: Box-Supervised 3D Point Cloud Instance Segmentation Using Gaussian Processes as Pseudo Labelers

Tuan Duc Ngo, Binh-Son Hua, Khoi Nguyen

ICCV 2023 · Point clouds › Segmentation and detection · approach: Point cloud semantic segmentation · 5 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 GP-generated pseudo labels instead of ignoring overlapped regions (ablation)ISBNet box-supervised · ScanNetV2 valAP: 38.1 -> 48.9Improved
use GP instead of assign-to-smaller-box heuristic (ablation)pseudo labeling · ScanNetV2 valAP: 41.8 -> 48.9Improved
use GP classification rather than linear classifier (ablation)pseudo labeling · ScanNetV2 valAP: 44.2 -> 48.9Improved
run GP on superpoints rather than points (ablation)GaPro · ScanNetV2 valAP: 45.7 -> 48.9Improved
GP classification instead of GP regression (ablation)GaPro · ScanNetV2 valAP: 47.8 -> 48.9Improved
add KL divergence loss on mean and variance (ablation)GaPro + ISBNet · ScanNetV2 valAP: 48.9 -> 49.2Improved
add self-training with deep features (ablation)GaPro + ISBNet · ScanNetV2 valAP: 48.9 -> 50.0Improved
use MSE loss for uncertainty instead of no loss (ablation)GaPro + ISBNet · ScanNetV2 valAP: 50.0 -> 49.9No clear effect
learnable GP length and output scale (ablation)GaPro · ScanNetV2 valAP: 48.0 -> 50.6Improved
use deep features in GP instead of position (ablation)GaPro · ScanNetV2 valAP: 48.5 -> 50.6Improved
add noise to GT box corners (ablation)GaPro + ISBNet · ScanNetV2 valAP: 48.3 (2cm) -> 21.0 (20% dim)Hurt
drop GT boxes (ablation)GaPro + ISBNet · ScanNetV2 valAP: 49.6 (5%) -> 30.6 (80%)Hurt

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