SwiftVGGT: A Scalable Visual Geometry Grounded Transformer for Large-Scale Scenes
By: Jungho Lee , Minhyeok Lee , Sunghun Yang and more
Potential Business Impact:
Builds detailed 3D maps much faster.
3D reconstruction in large-scale scenes is a fundamental task in 3D perception, but the inherent trade-off between accuracy and computational efficiency remains a significant challenge. Existing methods either prioritize speed and produce low-quality results, or achieve high-quality reconstruction at the cost of slow inference times. In this paper, we propose SwiftVGGT, a training-free method that significantly reduce inference time while preserving high-quality dense 3D reconstruction. To maintain global consistency in large-scale scenes, SwiftVGGT performs loop closure without relying on the external Visual Place Recognition (VPR) model. This removes redundant computation and enables accurate reconstruction over kilometer-scale environments. Furthermore, we propose a simple yet effective point sampling method to align neighboring chunks using a single Sim(3)-based Singular Value Decomposition (SVD) step. This eliminates the need for the Iteratively Reweighted Least Squares (IRLS) optimization commonly used in prior work, leading to substantial speed-ups. We evaluate SwiftVGGT on multiple datasets and show that it achieves state-of-the-art reconstruction quality while requiring only 33% of the inference time of recent VGGT-based large-scale reconstruction approaches.
Similar Papers
FlashVGGT: Efficient and Scalable Visual Geometry Transformers with Compressed Descriptor Attention
CV and Pattern Recognition
Makes 3D pictures from many photos faster.
Building temporally coherent 3D maps with VGGT for memory-efficient Semantic SLAM
CV and Pattern Recognition
Helps robots see and understand moving things.
LiteVGGT: Boosting Vanilla VGGT via Geometry-aware Cached Token Merging
CV and Pattern Recognition
Makes 3D pictures from many photos faster.