Preprint
Aug 2026
WALDO: One-Shot Exemplar-Conditioned Object Detection in Cluttered Scenes
WALDO is presented, a one-shot exemplar- and language-conditioned detection head with 3.4M trainable parameters that reads frozen V-JEPA 2.1 features to jointly predict object localization and target presence, with no gradient on the backbone.
K. Gupta, Ahmed Rafi Hasan, Md. Mahfuzur Rahman et al.
· 0 citations