mrharicot/monodepth

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

2265 stars · 625 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A TensorFlow implementation of unsupervised monocular depth estimation from single images using convolutional neural networks, based on the CVPR 2017 paper 'Unsupervised Monocular Depth Estimation with Left-Right Consistency'. It includes pretrained models for KITTI and Cityscapes and scripts for training on rectified stereo pairs.

Use cases

• predict depth maps from single images
• train a self-supervised depth estimation model on stereo pairs
• run pretrained monodepth models on KITTI or Cityscapes images
• research monocular depth estimation with left-right consistency
• estimate scene depth for autonomous driving datasets

When to choose

• you need a research-grade unsupervised depth estimation baseline
• you have rectified stereo training data like KITTI or Cityscapes
• you want to reproduce or build on the CVPR 2017 monodepth paper

When to avoid

• you need state-of-the-art accuracy - use monodepth2 instead
• you need modern TensorFlow 2.x or PyTorch support
• you lack stereo training data or a GPU for training

Facets

library · maturity maintenance

machine-learning deep-learning computer-vision image-processing computer-vision deep-learning machine-learning autonomous-vehicles python depth-estimation monocular-depth self-supervised-learning tensorflow stereo-vision cvpr-2017 research-code linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="mrharicot/monodepth")