Learning 3D Editing without Paired Supervision via Generative Prior Distillation

Researchers propose a method for instruction-guided 3D editing that eliminates the need for paired supervision by distilling generative priors, addressing the scarcity of high-quality training data. The approach aims to overcome the limitations of existing methods that rely on slow test-time optimization or pseudo-pairs.

Instruction-guided 3D editing is essential for interactive content creation, yet it faces a significant bottleneck: the severe scarcity of high-quality paired training data. Existing approaches attempt to bypass this by either relying on slow test-time optimization or training on pseudo-pairs constr