#212 · Primary category： Computer Vision
MonoScene
[CVPR 2022] "MonoScene: Monocular 3D Semantic Scene Completion": 3D Semantic Occupancy Prediction from a single image
Project last updated：03/25/26
GitHub Stars
819
Forks
78
Contributors
1
License
Apache-2.0
Why we included this project
MonoScene is a solid reference point if you work on 3D perception for autonomous driving or robotics. It takes a single monocular RGB image and reconstructs a dense 3D semantic occupancy grid, predicting both the geometry and the semantic class of every voxel in the scene. The repo, published with the CVPR 2022 paper from Inria, includes training and evaluation pipelines for SemanticKITTI and NYUv2, pretrained checkpoints, and a live Hugging Face demo, so you can reproduce the reported results or use the code as a baseline for your own occupancy prediction work. The README also points to related camera-only 3D occupancy projects, which makes it a handy entry point into that line of research.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
opencv
Open Source Computer Vision Library
RuView
π RuView turns commodity WiFi signals into real-time spatial intelligence, vital sign monitoring, and presence detection — all without a single pixel of video.
PaddleOCR
Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.
MinerU
Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.
tesseract
Tesseract Open Source OCR Engine (main repository)