cross-modal feature mapping

transform

PointCloud -> FeatureMap

Map multi-modal inputs like 3D point clouds to a shared latent space of a pre-trained visual model to leverage 2D semantic priors.

Problem it solves

Unsupervised 3D anomaly detection suffers from a lack of pre-trained models comparable in scale to 2D foundation models.

Consumes

PointCloud

Emits

FeatureMap

Distilled from 1 source

The real projects this mechanism was found in. Attribution is the point — this is how the best teams actually do it.