Changchun Zhou

IncVGGT: Incremental VGGT for Memory-Bounded Long-Range 3D Reconstruction

We present IncVGGT, a training-free incremental variant of VGGT that makes transformer-based 3D reconstruction feasible for long sequences in real-world applications. Vanilla VGGT relies on dense global attention, which causes memory to grow quadratically and requires excessive computation, making i…