Skip to content
conftrace_

visual geometry grounded transformer

2 papers

Explore in graph

Also known as

VGGT

Co-occurring keywords · papers mentioning both

fp8 quantization (1) geometric perception (1) depth map estimation (1) point cloud prediction (1) token merging (1) camera pose estimation (1) mixed-precision quantization (1) vision foundation model (1) 3d reconstruction (1) post-training quantization (1)

Papers

LiteVGGT: Boosting Vanilla VGGT via Geometry-aware Cached Token Merging CVPR 2026 QVGGT: Post-Training Quantized Visual Geometry Grounded Transformer CVPR 2026