xheon/panoptic-reconstruction
Official implementation of the NeurIPS 2021 paper "Panoptic 3D Scene Reconstruction from a Single RGB Image"
This project helps professionals working with 3D environments, such as robotics engineers, augmented reality developers, or architects. It takes a single 2D image as input and produces a detailed 3D reconstruction of the scene, identifying individual objects and classifying them by type. This allows users to understand the geometry and contents of a 3D space from just one photo.
210 stars. No commits in the last 6 months.
Use this if you need to quickly generate a comprehensive 3D model, complete with object identification and semantic labels, from a single photograph of an indoor scene.
Not ideal if your primary goal is only basic geometric reconstruction without semantic or instance segmentation, or if you are working with outdoor scenes.
Stars
210
Forks
26
Language
Python
License
—
Category
Last pushed
Feb 18, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/xheon/panoptic-reconstruction"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
3DOM-FBK/deep-image-matching
Multiview matching with deep-learning and hand-crafted local features for COLMAP and other SfM...
suhangpro/mvcnn
Multi-view CNN (MVCNN) for shape recognition
zouchuhang/LayoutNet
Torch implementation of our CVPR 18 paper: "LayoutNet: Reconstructing the 3D Room Layout from a...
andyzeng/tsdf-fusion-python
Python code to fuse multiple RGB-D images into a TSDF voxel volume.
andyzeng/tsdf-fusion
Fuse multiple depth frames into a TSDF voxel volume.