This interactive application utilizes quantized edge models derived from the groundbreaking MoGe monocular geometry estimation series developed by Ruicheng Wang et al. (USTC, Microsoft Research) and deployed via Google LiteRT:
1. MoGe-2: Accurate Monocular Geometry with Metric Scale & Sharp Details
@article{wang2025moge2,
title={MoGe-2: Accurate Monocular Geometry with Metric Scale and Sharp Details},
author={Wang, Ruicheng and Peng, Sida and Guo, Chen and others},
journal={arXiv preprint arXiv:2507.02546},
year={2025}
}
2. MoGe (v1): Unlocking Accurate Monocular Geometry for Open-Domain Images
Authors: Ruicheng Wang, Shengyi Qian, Christian Rupprecht, David F. Fouhey Publication: IEEE/CVF Computer Vision and Pattern Recognition (CVPR 2024 / arXiv:2410.19115)
@inproceedings{wang2024moge,
title={MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision},
author={Wang, Ruicheng and Qian, Shengyi and Rupprecht, Christian and Fouhey, David F.},
booktitle={CVPR},
year={2024}
}
3. LiteRT Quantization & Hugging Face Edge Checkpoints