Concept
Geometric foundation model
A geometric foundation model is a large pretrained network that infers 3D structure (depth, camera pose, point maps, surface normals) directly from ordinary 2D images. Think of it like an image classifier backbone such as ResNet or a ViT, except the pretraining objective is…
The rest of “Geometric foundation model” is a premium feature: every concept in the library gets a precise, practitioner-focused write-up like this one, cross-linked straight from the paper summaries that use it.
Log in to unlock→