Bekci/ESCAPE
ESCAPE: Equivariant Shape Completion via Anchor Point Encoding
This project helps computer vision researchers and robotics engineers reconstruct complete 3D object shapes from incomplete or partial 3D scan data, regardless of how the object is oriented. You feed it a partial 3D point cloud, and it outputs a highly accurate and complete 3D shape, robust to rotations and translations. It's designed for professionals working with 3D perception in robotics, augmented reality, or 3D modeling.
Use this if you need to reliably reconstruct full 3D object geometries from fragmented 3D sensor data where object orientation might vary.
Not ideal if your task involves 2D image processing or if you only need object classification without 3D shape reconstruction.
Stars
9
Forks
—
Language
Python
License
—
Category
Last pushed
Nov 19, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/Bekci/ESCAPE"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
cvg/LightGlue
LightGlue: Local Feature Matching at Light Speed (ICCV 2023)
shariqfarooq123/AdaBins
Official implementation of Adabins: Depth Estimation using adaptive bins
amusi/CVPR2026-Papers-with-Code
CVPR 2026 论文和开源项目合集
isl-org/ZoeDepth
Metric depth estimation from a single image
daviddmc/NeSVoR
NeSVoR is a package for GPU-accelerated slice-to-volume reconstruction.