#190 · Primary category： Computer Vision
ScanNet
ScanNet is a large-scale RGB-D video dataset with 2.5 million views in 1500+ scans, offering 3D poses, reconstructions, and instance-level semantic annotations.
Project last updated：11/03/25
GitHub Stars
2.3K
Forks
374
Contributors
6
License
Other
Why we included this project
ScanNet has been the default benchmark for indoor 3D scene understanding for years. The dataset packs 2.5 million RGB-D views into more than 1,500 scans, each with camera poses, surface reconstructions, and instance-level semantic labels. Because this repository is the official hub for the data, it is also where researchers go to reproduce published results or compare a new model against a widely cited baseline. Per-scan folders bundle the reconstructed meshes, over-segmentations, and 2D label projections together, so setting up training or evaluation on semantic segmentation and scene reconstruction is mostly a matter of downloading and unpacking.
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)