#458 · Primary category： Computer Vision
CREStereo
Official MegEngine implementation of CREStereo(CVPR 2022 Oral).
Project last updated：02/10/23
GitHub Stars
635
Forks
68
Contributors
5
License
Apache-2.0
Why we included this project
CREStereo is a stereo-matching network that estimates per-pixel disparity from a pair of left and right images, using a cascaded recurrent architecture with adaptive correlation. The repository is the official MegEngine implementation of the CVPR 2022 paper, and it includes pretrained weights, a one-command inference script that outputs a disparity map, and training configuration files. It also ships a large proposed training dataset and download scripts for standard benchmarks such as SceneFlow, KITTI, and Middlebury, so you can reproduce results or fine-tune on your own stereo pairs. If you work on robotics, autonomous driving, or 3D reconstruction, the practical inference path and the provided Docker image make it easy to get a working disparity pipeline up and running.
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)