MegBA: A High-Performance and Distributed Library for Large-Scale Bundle Adjustment

12/02/2021
by   Jie Ren, et al.
0

Large-scale Bundle Adjustment (BA) is the key for many 3D vision applications (e.g., Structure-from-Motion and SLAM). Though important, large-scale BA is still poorly supported by existing BA libraries (e.g., Ceres and g2o). These libraries under-utilise accelerators (i.e., GPUs), and they lack algorithms to distribute BA computation constrained by the memory on a single device. In this paper, we propose MegBA, a high-performance and distributed library for large-scale BA. MegBA has a novel end-to-end vectorised BA algorithm that can fully exploit the massive parallel cores on GPUs, thus speeding up the entire BA computation. It also has a novel distributed BA algorithm that can automatically partition BA problems, and solve BA sub-problems using distributed GPUs. The GPUs synchronise intermediate solving state using network-efficient collective communication, and the synchronisation is designed to minimise communication cost. MegBA has a memory-efficient GPU runtime and exposes g2o-compatible APIs. Experiments show that MegBA can out-perform state-of-the-art BA libraries (i.e., Ceres and DeepLM) by up to 33x and 3.3x respectively, in public large-scale BA benchmarks. The code of MegBA is available at: <https://github.com/MegviiRobot/MegBA>.

READ FULL TEXT
research
03/13/2018

CuLDA_CGS: Solving Large-scale LDA Problems on GPUs

Latent Dirichlet Allocation(LDA) is a popular topic model. Given the fac...
research
03/20/2022

Exascale Grid Optimization (ExaGO) toolkit: An open-source high-performance package for solving large-scale grid optimization problems

This paper introduces the Exascale Grid Optimization (ExaGO) toolkit, a ...
research
05/11/2023

Decentralization and Acceleration Enables Large-Scale Bundle Adjustment

Scaling to arbitrarily large bundle adjustment problems requires data an...
research
08/02/2020

Stochastic Bundle Adjustment for Efficient and Scalable 3D Reconstruction

Current bundle adjustment solvers such as the Levenberg-Marquardt (LM) a...
research
03/24/2021

FastMoE: A Fast Mixture-of-Expert Training System

Mixture-of-Expert (MoE) presents a strong potential in enlarging the siz...
research
07/17/2023

Distributed bundle adjustment with block-based sparse matrix compression for super large scale datasets

We propose a distributed bundle adjustment (DBA) method using the exact ...
research
08/24/2020

CRAC: Checkpoint-Restart Architecture for CUDA with Streams and UVM

The share of the top 500 supercomputers with NVIDIA GPUs is now over 25 ...

Please sign up or login with your details

Forgot password? Click here to reset