Hello,
I would like to ask if it is possible to integrate COLMAP in the monocular video process and, if yes, how difficult it would be. I'm using monocular input, and because of this, only the front view of the constructed avatar can be represented from the Gaussian splats. I need to reconstruct an avatar that can be viewed from different angles, as it was reconstructed with multiview videos, that's why I'm asking this. Thank you for your time!
Hello,
I would like to ask if it is possible to integrate COLMAP in the monocular video process and, if yes, how difficult it would be. I'm using monocular input, and because of this, only the front view of the constructed avatar can be represented from the Gaussian splats. I need to reconstruct an avatar that can be viewed from different angles, as it was reconstructed with multiview videos, that's why I'm asking this. Thank you for your time!