SOURCE-LINKED INTELLIGENCE
DDMS: Discriminative Distillation of Multi-view Foundational Features into Single-view Models
Foundational visual features such as DINO have played a critical role across modern computer vision, and have recently become key components in multi-view feed-forward geometry estimators. In this work, we demonstrate that by re-distilling these multi-view models---their internal knowledge of 3D geometry---into a single-view estimator, we can obtain enhanced 3D consistent foundational features. Our key idea is to construct a multi-view teacher by fusing pretrained 2D foundation features with multi-view geometric features, and refining the fused representation with a discriminative ranking obje
Read original source ↗ Open in workspace
- recordType
- paper
- region
- Global
Evidence & attribution
- arXiv · AI, language, vision and robotics · 2026-08-24T21:38:49.000Z
First collected: 2026-09-21T10:22:00.206Z. This is not the publication date.