AI Paper Atlas

Research map / Learning theory and algorithms / Matrix and tensor methods

Distributed Bundle Adjustment with Block-Based Sparse Matrix Compression for Super Large Scale Datasets

Maoteng Zheng, Nengcheng Chen, Junfeng Zhu, Xiaoru Zeng, Huanbin Qiu, Yuyao Jiang, Xingyue Lu, Hao Qu

ICCV 2023 · Learning theory and algorithms › Matrix and tensor methods · approach: Randomized sketching for matrices · 6 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
distribute RCS formation and PCG across computing nodes with BSMC storageexact LM bundle adjustment · Finalmain node memory (GB): 16.8 -> 4.93 (vs Ceres-CG)Improved
use distributed CPU nodes instead of Ceres-CGexact LM bundle adjustment · Venicerun time (s): 1992 -> 69.7Improved
compare with GPU-based MegBA on small public datasetsOurs · Ladybug/Venice/Finalrun time (s): Final: 906 vs 22.6 (MegBA)Hurt
use BSMC format instead of CSR for the reduced camera system (ablation)RCS storage · memorymemory requirement: 1.5 to 3.0 times smallerImproved
increase number of computing nodes (2, 6, 10) (ablation)Ours · SW, DaPu, GDSefficiencyImproved
vary parallel task number (ablation)Ours · SW, DaPu, GDSefficiencyNo clear effect
apply parallel mode to small datasets (under 2K images)Ours · small datasetsefficiencyNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy