arXiv AI

Think Locally, Refine Globally for Memory-Efficient 3D Reconstruction

LoG-VGGT is a memory‑efficient framework for long‑sequence 3D reconstruction that balances local temporal modeling with global camera consistency. It uses cross‑window attention in a small subset of transformer blocks to propagate information across adjacent temporal windows while keeping memory usage bounded. A global camera consistency refinement module further improves long‑horizon pose stability by enforcing scene‑level constraints through cross‑attention between camera and compact register tokens, leading to better depth accuracy and robust camera pose estimation on multiple benchmarks.

arXiv Computer Vision
Oct 23

Stream4D: Streaming 4D Reconstruction with Training-free Motion Decoupling via Causal KV-Cache Attention

Stream4D is a training‑free, motion‑decoupled streaming 4D reconstruction framework that can be integrated into causal Transformer backbones such as StreamVGGT and Stream3R. It exploits the causal attention’s KV cache to aggregate temporal motion cues, fuses them with depth priors, and produces precise dynamic masks that separate static and dynamic regions. A soft key masking scheme then suppresses dynamic interference in attention, improving camera pose estimation and reducing ghosting in long sequences.

By Wei Zhang, Ying Feng, Mengtong Wu, Shiqiang Gong, Huaqi Zhang, Qi Zhang
arXiv Computer Vision
Sep 15

FFVO: A Feedforward Pose Decoder for Long-Horizon Visual Odometry

arXiv:2609.13733v1 Announce Type: new Abstract: Stable and reliable 4D spatial understanding is fundamental for autonomous driving systems. While feedforward reconstruction networks can estimate came...

By Meng-Li Shih, Shih-Yang Su, Yuliang Zou, Hao Xiang, Haidong Zhu, Vincent Casser, Brian Curless, Dmitry Kalenichenko, Mingxing Tan, Dragomir Anguelov
arXiv Computer Vision
Sep 15

Anchor3R: Streaming 3D Reconstruction with Transient Anchors for Long-Horizon Visual Mapping

Anchor3R is a streaming 3D reconstruction framework that predicts window-relative poses and local geometry in the current‑frame coordinate system, forming a dense relative‑pose graph for online pose updates and loop‑aware motion averaging. It improves long‑horizon pose accuracy and dense reconstruction quality on indoor, outdoor, driving, and RGB‑D benchmarks, and generalizes from 48‑frame training sequences to streams exceeding 10,000 frames while keeping GPU memory bounded. The method addresses issues of train‑test mismatch, early‑anchor bias, and accumulated drift found in previous streaming models.

By Peilin Tao, Chong Cheng, Yuansen Du, Caiwei Song, Zhengqing Chen, Xiaoyang Guo, Wei Yin, Weiqiang Ren, Qian Zhang, Hainan Cui, Shuhan Shen
arXiv Computer Vision
Oct 22

BlazeVGGT: 13x Faster VGGT at 2,000 Frames via Camera Token without Training

BlazeVGGT is a training‑free, plug‑and‑play module that speeds up the VGGT visual geometry model by sparsifying cross‑view attention. It selects a compact set of complementary views using a greedy determinantal point process, preserving dense per‑frame predictions while eliminating redundant computations. On 2,000‑frame sequences, BlazeVGGT achieves roughly 13× faster inference and improves Chamfer Distance from 0.41 to 0.32, demonstrating a strong accuracy–efficiency trade‑off for point‑cloud reconstruction and camera‑pose estimation.

By Shujun Yao, Junhui Lin, Congying Han, Tiande Guo, Ziwen Liu, Bonan Li
arXiv Computer Vision
Nov 5

RegimeVGGT: Layer-Wise Spatially Preserving Redundancy Removal for Visual Geometry Grounded Transformer

RegimeVGGT introduces a layer‑wise compression strategy for the Visual Geometry Grounded Transformer (VGGT) that preserves spatial structure while reducing redundancy. By analyzing spectral, probing, and causal properties, the authors identify three functional regimes—shallow, middle, and deep layers—and apply a U‑shaped compression that protects salient tokens and cross‑frame spatial coverage. The method achieves a 6.7× speedup over the memory‑optimized VGGT* without any training, maintaining comparable reconstruction quality.

By Jinhao You (University of Pennsylvania), Shuo Lyu (University of Pennsylvania), Zhuohang Lyu (University of Pennsylvania), Tanxuan Li (University of Pennsylvania), Zibo Zhao (University of Pennsylvania), Jiaxiang Hu (University of California, Irvine), Kai Tang (Nanyang Technological University), Yichen Guo (Nanyang Technological University)
Hugging Face Trending Papers
Jun 2

BA-T: An Iterative Transformer for Two-View Bundle Adjustment

Feed-forward models for 3D reconstruction have achieved strong performance using deep cross-view attention to exchange information across images. However, these approaches often depend on heavy decoder stacks and lack a structured mechanism for geometry refinement, resulting in poor multi-view consistency.