lpiccinelli-eth/UniDepth
A deep learning model for estimating metric depth from single RGB images, supporting any camera setup without retraining.
Not currently ranked — collecting fresh signals.
star history
UniDepth provides universal monocular metric depth estimation by training a deep neural network to predict metric depth from a single image without requiring camera calibration. The model can generalize to any camera without domain-specific fine-tuning. It achieves top results on established benchmarks like KITTI and NYU Depth V2. The project includes training code, evaluation scripts, and pre-trained model weights.
Frequently asked
- What is lpiccinelli-eth/UniDepth?
- A deep learning model for estimating metric depth from single RGB images, supporting any camera setup without retraining.
- Is UniDepth open source?
- Yes — lpiccinelli-eth/UniDepth is an open-source project tracked on heatdrop.
- What language is UniDepth written in?
- lpiccinelli-eth/UniDepth is primarily written in Python.
- How popular is UniDepth?
- lpiccinelli-eth/UniDepth has 1.2k stars on GitHub.
- Where can I find UniDepth?
- lpiccinelli-eth/UniDepth is on GitHub at https://github.com/lpiccinelli-eth/UniDepth.