AI Paper Atlas

Research map / Point clouds / Shape and surface completion

USIP: Unsupervised Stable Interest Point Detection From 3D Point Clouds

Jiaxin Li, Gim Hee Lee

ICCV 2019 · Point clouds › Shape and surface completion · approach: Point cloud upsampling and normals · 215 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 USIP detector instead of 3DFeatNet detectorOur Desc. on KITTI registration · KITTI, 256 keypointsregistration failure rate (%): 2.61 -> 1.41Improved
use USIP detector instead of 3DFeatNet detectorFPFH on KITTI registration · KITTI, 256 keypointsregistration failure rate (%): 12.15 -> 8.37Improved
use USIP detector instead of 3DFeatNet detectorSHOT on KITTI registration · KITTI, 256 keypointsregistration failure rate (%): 11.76 -> 5.40Improved
use USIP detector instead of 3DFeatNet detectorOur Desc. on KITTI registration · KITTI, 256 keypointsinlier ratio (%): 15.66 -> 32.20Improved
use USIP detector in optimal configuration instead of 3DFeat-Netregistration pipeline · KITTIregistration failure rate (%): 0.57 -> 0.24Improved
use USIP detector instead of 3DFeatNet detector for speedkeypoint extraction · KITTI 128 keypointstime (s): 0.438 -> 0.011Improved
train descriptor with USIP keypoints and saliency uncertainty instead of random keypoints (ablation)Our Desc. · KITTI registrationregistration failure rate (%): 1.02 -> 0.24Improved
train descriptor with USIP keypoints and saliency uncertainty instead of random keypoints (ablation)Our Desc. · Oxford registrationregistration failure rate (%): 1.20 -> 0.93Improved
increase number of neighbors K in the FPN (larger receptive field) (ablation)USIP FPN · ModelNet40 (M=64)keypoint degeneracy: K=9 no degeneracy; K=24 principal axis; K=64 centroidHurt
use too small number of proposals M or too large K (ablation)USIP · point cloud registrationregistration failure rate (%)Hurt
add point-to-point loss (lambda=6 vs lambda=0) (ablation)USIP · repeatability, 128 keypointsrelative repeatabilityNo clear effect
use large lambda for point-to-point loss (ablation)USIP · training convergenceconvergenceHurt

Later papers that compared with this method

2 later accepted papers report USIP as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

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