CONCAVITY-BASED GROUPING FOR UNORGANIZED 3D POINT CLOUD SEGMENTATION AND ABSTRACTION
Patent №
US 12,688,585
Granted
2026-07-21
Filed 2024
Owner
NEW YORK UNIVERSITY
Lab
—
AI components
0
Assignment
None on record
Dataset
AIPD
Application
18709290
Points are grouped from a point cloud using Concavity-induced Distance (CID). First, a set of seed points is sampled from the input point cloud. Then a grouping of points is conducted based on a nearest neighbor search computed using CID. These two steps enable two novel solutions in point cloud segmentation and scene abstraction. In one embodiment, CID is determined between two points residing within an object surface. In a second embodiment, a CID is determined between groups of points.
G06T 7/12G06T 7/11G06T 7/187G06T 7/64G06V 10/806G06V 20/64G06V 10/82G06V 10/26+6 more
Ownership
NEW YORK UNIVERSITY
From the same owner
SYSTEM, METHOD AND COMPUTER-ACCESSIBLE MEDIUM FOR DECORRELATING AND REMOVING NOISE, AND/OR REMOVING PARTIAL FOURIER-INDUCED GIBBS (RPG) RINGING ARTIFACTS FROM AT LEAST ONE IMAGEUS 12,632,935 · 2026System and Method for Robotic InspectionUS 12,591,041 · 2026SYSTEM TO DETECT DIELECTRIC CHANGES IN MATTERUS 12,567,503 · 2026SYSTEMS, METHODS AND COMPUTER-ACCESSIBLE MEDIUM FOR IDENTIFYING TARGET PAIRS FOR CAR-T THERAPYUS 12,562,256 · 2026SYSTEM AND METHOD FOR TUMOR CHARACTERIZATIONUS 12,536,223 · 2026ANOMALY DETECTION IN REAL-TIME MULTI-THREADED PROCESSES ON EMBEDDED SYSTEMS AND DEVICES USING HARDWARE PERFORMANCE COUNTERS AND/OR STACK TRACESUS 12,450,353 · 2025DISPLAY GUIDED HIGH-ACCURACY ROBOTIC NAVIGATION AND MOTION CONTROL SYSTEM AND METHODSUS 12,384,115 · 2025DETERMINING CHARACTERISTICS OF MUSCLE STRUCTURES USING ARTIFICIAL NEURAL NETWORKUS 12,354,730 · 2025