cs.CVNov 13, 2026

Eco3R: Pinned-Key Inference for Efficient and Streamable Feed-Forward 3D Reconstruction from Long Videos

Authors: Hongsen Cao, Shanxin Yuan, Mona Jaber, Ahmed Sayed

Organizations: School of Electronic Engineering and Computer Science, Queen Mary University of London, London, UK

Abstract

Feed-forward multi-view transformers such as VGGT recover the cameras and geometry of a video in one pass, but the cost of their cross-frame self-attention grows quadratically with the frame count, and their accuracy drops on long videos. Most remedies approximate attention over all frames under a fixed computational budget or select a few keys by their content. We hypothesize that the keys of a long video can be selected independently of video content, and our experiments on VGGT and π3π^3 support it: a small number of uniformly sampled frames can serve as the keys of every frame. We present Eco3R, a training-free inference framework whose keys form a Pinned Key Set of 31 such frames. A pin pass stores their tokens in a Scene Cache, and the other frames read it in independent batches that reproduce joint inference with these keys exactly, at a constant cost per frame. An Online Stream updates the pinned frames during recording. Six middle global blocks align the frames and form a reusable VGGT module: the Alignment Score reads their attention to flag failures, and Alignment Refocusing reduces them by fine-tuning only these blocks (3.6 GPU hours). Eco3R (π3π^3) matches π3π^3's success rate on 1,000-frame ScanNet videos while running 13.8 times faster. With refocusing, Eco3R (VGGT) raises VGGT's success rate from 70% to 82% while running 8.5 times faster on 500 frames. Eco3R can reconstruct sequences of up to 4,000 frames on a single A100 GPU, while Eco3R (π3π^3) streams 1,000 frames at 18.6 fps, achieving the lowest online absolute trajectory error.

Figures & tables

Explore similar work

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

    Oct 21, 2026Shujun Yao, Junhui Lin, Congying Han +3Multi-View 3D ReconstructionFeed-Forward 3D Reconstruction

  2. Attention Itself Could Retrieve.RetrieveVGGT: Training-Free Long Context Streaming 3D Reconstruction via Query-Key Similarity Retrieval

    May 10, 2026Zichen Zou, Xiaosong Jia, Zuxuan Wu +1Memory-Efficient Attention3D ViTs

  3. OVGGT: O(1) Constant-Cost Streaming Visual Geometry Transformer

    Mar 6, 2026Si-Yu Lu, Po-Ting Chen, Hui-Che Hsu +3Streaming 3D Reconstruction3D ViTs