MoCha is a pioneering model for Dialogue-driven Movie Shot Generation.
| 🌐Project Page | 📖Paper | 🔗Github | 🤗Demo|
We introduce our evaluation benchmark "MoChaBench", as described in Section 4.3 of the MoCha Paper.
MoChaBench is tailored for Dialogue-driven Movie Shot Generation — generating movie shots from a combination of speech and text(speech + text → video).
It complements existing narration-style, non-dialogue scene generation benchmarks (text → video), such as… See the full description on the dataset page:
https://huggingface.co/datasets/CongWei1230/MoChaBench.
