Closed novwaul closed 10 months ago
Do I have to train each separate model per object? Or is it enough to train only one model to synthesize new scenes for arbitrary objects?
Our model is a per-scene optimization approach. Currently, we are not generalizable to different scenes.
Do I have to train each separate model per object? Or is it enough to train only one model to synthesize new scenes for arbitrary objects?