4 papers
Two Global Crops Suffice: Locating Semantic Emergence in DINO-Style Self-Supervised Learning
Basavaraj Sunagad, Artur Jesslen, Adam Kortylewski
Self-supervised vision transformers trained with DINO-style objectives exhibit striking emergent semantic representation quality across visual tasks, yet the mechanisms underlying…
Category-Level 3D Correspondence in Camera Space via Morphable Object Priors
Leonhard Sommer, Artur Jesslen, Basavaraj Sunagad +1
Understanding 3D objects from images is fundamental to robotics and AR/VR applications. While recent work has made progress in category-level pose estimation, current representatio…
SOCO: Benchmarking Semantic Object Correspondence in Vision Foundation Models
Olaf Dünkel, Basavaraj Sunagad, Haoran Wang +3
Measuring structured object understanding in vision foundation models remains challenging due to inconsistent evaluation protocols and limited part-level supervision. Semantic corr…
TEDRA: Text-based Editing of Dynamic and Photoreal Actors
Basavaraj Sunagad, Heming Zhu, Mohit Mendiratta +3
Over the past years, significant progress has been made in creating photorealistic and drivable 3D avatars solely from videos of real humans. However, a core remaining challenge is…