What can Off-the-Shelves Large Multi-Modal Models do for Dynamic Scene Graph Generation?

Re-assign community