Closed Colezwhy closed 2 weeks ago
In the technical report of SAM2Point, the 3D volume is treated as video frames when segmenting 3D point clouds. However, I come up with some questions. Why design like this? What does it mean and why it can work? Thank you for the interesting work!
In the technical report of SAM2Point, the 3D volume is treated as video frames when segmenting 3D point clouds. However, I come up with some questions. Why design like this? What does it mean and why it can work? Thank you for the interesting work!