AI Paper Atlas

Research map / Point clouds / Segmentation and detection

SAM-guided Graph Cut for 3D Instance Segmentation

Haoyu Guo, He Zhu, Sida Peng, Yuang Wang, Yujun Shen, Ruizhen Hu, Xiaowei Zhou

ECCV 2024 · Point clouds › Segmentation and detection · approach: Scene completion and instance segmentation · 31 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 GNN refinement on top of SAM-annotated graph (ablation)SAM-guided graph cut · ScanNet200mAP: 19.7 -> 22.1Improved
add SAM-derived edge weights to the graph (ablation)SAM-guided graph cut · ScanNet200mAP: 10.1 -> 22.1Improved
add SAM node features (ablation)SAM-guided graph cut · ScanNet200mAP: 19.4 -> 22.1Improved
add regularization loss for edges outside pseudo-labels (ablation)GNN training · ScanNet200mAP: 19.9 -> 22.1Improved
increase prompt points k from 1 to 5 (ablation)SAM-guided graph cut · ScanNet200mAP: 20.3 -> 22.1Improved
increase prompt points k from 5 to 9 (ablation)SAM-guided graph cut · ScanNet200mAP: 22.1 -> 21.9Hurt
increase min co-visible views n from 10 to 30 (ablation)pseudo-label generation · ScanNet200mAP: 22.1 -> 20.2Hurt
use CropFormer instead of SAM to generate pseudo-labels (ablation)GNN training · ScanNet200 pseudo-label qualityaccuracy: 0.892 -> 0.912Improved
use direct graph partition (union-find) instead of NCuts or DBSCAN (ablation)SAM-annotated graph · ScanNet200mAP: 15.7 -> 19.7 (NCuts); 10.3 -> 19.7 (DBSCAN)Improved
train GNN on ScanNet200 pseudo-labels, test on other datasets (ablation)SAM-guided graph cut · KITTI-360mAP: 22.6 -> 23.8Improved
use supervised Mask3D instead of the pseudo-label method on out-of-domain dataMask3D · KITTI-360mAP: 23.8 -> 0.2Hurt
add GNN to NCuts graph cut on KITTI-360 (ablation)SAM graph + NCuts · KITTI-360mAP: 19.5 -> 18.1Hurt

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