Scene-Graph ViT: End-to-End Open-Vocabulary Visual Relationship Detection.

Tim Salzmann, Markus Ryll, Alex Bewley, Matthias Minderer

Browse the full ECCV paper archive.