Mark Boss

Achieve precise, intuitive 3D control over video generation by moving rigid object meshes, eliminating depth ambiguity and viewpoint inconsistencies.

A single video tokenizer can achieve superior semantic alignment and fidelity compared to much larger models by directly conditioning on frozen visual features.

Constraint meshes in Arbor provide a powerful new way to dictate 3D object placement and interaction, significantly improving control over asset generation.

Fine-tuning image layer decomposition models without paired supervision leads to significantly improved layer quality and reduced artifacts.