cs.CVNov 7, 2026

PoseBench: Probing Pose Awareness in Visual Foundation Models

Authors: Most Nilufa Yeasmin, Draix Wyatt, Ruyi Lian

Organizations: South Dakota State University, Brookings, SD, USA

Abstract

Pose awareness characterizes the extent to which a learned feature space aligns with the underlying object-pose manifold, such that observations with similar poses exhibit greater feature similarity than those with increasingly different poses. While such structure is desirable for pose-sensitive 3D vision tasks, visual foundation models pre-trained on individual 2D images typically lack explicit 3D or pose supervision. This raises a fundamental question: Do 2D visual foundation models exhibit pose awareness after large-scale pretraining? To answer this question, we introduce PoseBench, a systematic benchmark to probe pose awareness in pretrained visual features. Our evaluation follows a progressive design: held-out rendered views probe intrinsic pose-aware structure under controlled conditions, while real-image queries evaluate how robustly this structure remains accessible under real-world variations. We further introduce pose-graph-based refinement as a complementary probe of compatibility between feature neighborhoods and the underlying pose manifold. Across 31 publicly available checkpoints from 10 pretraining families, PoseBench reveals that pose awareness is an implicit but non-uniform property of 2D visual foundation models, and consistent gains from pose-graph refinement suggest that feature spaces retain structure compatible with the underlying pose manifold. PoseBench therefore provides a systematic basis for understanding and selecting pretrained features for pose-sensitive 3D vision tasks.

Figures & tables

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. SeeSE3: Emergence of 3D Space in Vision Features

    Jul 15, 2026Caroline Chen, Sayna Ebrahimi, Fedor Kitashov +4Geometric Representation LearningVisual Odometry

  2. Geometry Matters: 3D Foundation Priors for Learning Semantic Correspondence

    May 28, 2026Artur Jesslen, Olaf Dünkel, Adam KortylewskiSemantic Correspondence

  3. Lightweight 3D Feature Pretraining by Bayesian Inversion of 2D Foundation Models

    Jun 19, 2026Marwane Hariat, Gianni Franchi, David Filliat +1Multi-View ConsistencyBayesian Inference