Closed imbinwang closed 3 months ago
That’s a good question and currently the controllability of camera pose is not very accurate due to the lack of diverse training data. You can check the discussion here and Appendix J for some exploration.
Appendix J shows the camera fov could be edited. Thanks for clarification.
Another question: could the car be turned upside down to generate an accident scene by changing the orientation of the control box? The paper demonstrates the car can be rotated from head to tail. The nuscenes dataset also lacks such training cases.
I suppose not, because there are no such cases during training.
Get it. Thank you.
Thanks for sharing this work. As the section 4.1 of your paper shows, the camera parameters are encoded by Fourier embedding and a MLP. By this way, could we slightly change camera intrinsic/extrinsic to generate images? I didn't see any results in your paper about camera control.