Mate im not an AI expert but no one Gave a serious answer yet. it would take a very long time. Even a 5 sec video could take 5 minutes to generate with good hardware and that’s 1 version. Think about all the prompts and loras and other factors to tweak to get it right. Plus with the items will likely be inconsistent throughout the movie. Forcing it to have the same accessories for each character for consistency will be frustrating.
Maybe lower the scope to 1 still image with they are together in Riverdale.
Mate im not an AI expert but no one Gave a serious answer yet. it would take a very long time. Even a 5 sec video could take 5 minutes to generate with good hardware and that’s 1 version. Think about all the prompts and loras and other factors to tweak to get it right. Plus with the items will likely be inconsistent throughout the movie. Forcing it to have the same accessories for each character for consistency will be frustrating. Maybe lower the scope to 1 still image with they are together in Riverdale.