FoldingAgent converts origami demo videos into executable parametric folding programs

Maya Moriya · hf · 2026-09-03

FoldingAgent infers parametric origami procedures from demonstration videos: a vision-language model with specialized tools converts origami videos into executable parametric folding programs via sequential reasoning and physical verification.

The video-to-structured-program pipeline is relevant to robot manipulation and program synthesis.

Original post →

More from Embodied

Embodied channel →