الباحثون

Bardienus P. Duisterhof

المنشورات 2

نسخة أولية وصول مفتوح

PointZero: 3D Point Track Completion for Learning Transferable 3D Dynamics

Bardienus P. Duisterhof, Kaifeng Zhang, Adam Hung وآخرون · 2026

World models endow perceptual systems with the ability to predict how scenes evolve under interaction. They are most beneficial when trained on diverse volumes of data, to instill a rich prior into downstream applications. Existing methods typically require robot action labels to learn action-conditioned 3D dynamics, w …

نسخة أولية وصول مفتوح

Modality-Autoregressive World-Action Models

Adam Hung, Bardienus P. Duisterhof, Deva Ramanan وآخرون · 2026

World-action models (WAMs) jointly model future observations and actions, typically predicting the future as RGB images. Other visual modalities such as depth, pretrained visual features, and point tracks can more efficiently capture geometric, semantic, and motion features. However, how best to combine these modalitie …

المؤلفون المشاركون