Recast all three Wannabe dancers
Photo one is the front lead. Photo two is the green-gown dancer. Photo three is the gold-gown dancer. Separate images keep hair and likeness apart, so Wannabe AI does not melt three people into one dancer. Each uploaded outfit travels with that face. The Wannabe video only works when all three faces stay readable.