Get absolute depth of scene from one image in new domain by seperate camera from depth with UniDepth
Get absolute depth of scene from one image in new domain by seperate camera from depth with UniDepth
UniDepth: Universal Monocular Metric Depth Estimation
arXiv paper abstract https://arxiv.org/abs/2403.18913
arXiv PDF paper https://arxiv.org/pdf/2403.18913.pdf
Accurate monocular metric depth estimation (MMDE) is crucial ... However ... recent MMDE methods is confined to their training domains ... fail to generalize to unseen domains
... propose a new model, UniDepth, capable of reconstructing metric 3D scenes from solely single images across domains.
... UniDepth directly predicts metric 3D points from the input image at inference time without any additional information, striving for a universal and flexible MMDE solution.
... UniDepth implements a self-promptable camera module predicting dense camera representation to condition depth features.
... model exploits a pseudo-spherical output representation, which disentangles camera and depth representations ... propose a geometric invariance loss that promotes the invariance of camera-prompted depth features.
... demonstrate the superior performance of UniDepth, even when compared with methods directly trained on the testing domains ...
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comments