Decentralization and Acceleration Enables Large-Scale Bundle Adjustment

TL;DR

This paper introduces a fully decentralized approach for large-scale bundle adjustment that reduces computation and communication bottlenecks, enabling faster and more accurate solutions compared to centralized methods.

Contribution

The authors propose a novel surrogate function and optimization framework that allows parallel, decentralized bundle adjustment with theoretical convergence guarantees.

Findings

Achieves significant speedups over centralized methods (up to 953.7x).

Converges faster than decentralized baselines with similar resources.

Produces more accurate solutions than centralized baselines.

Abstract

Scaling to arbitrarily large bundle adjustment problems requires data and compute to be distributed across multiple devices. Centralized methods in prior works are only able to solve small or medium size problems due to overhead in computation and communication. In this paper, we present a fully decentralized method that alleviates computation and communication bottlenecks to solve arbitrarily large bundle adjustment problems. We achieve this by reformulating the reprojection error and deriving a novel surrogate function that decouples optimization variables from different devices. This function makes it possible to use majorization minimization techniques and reduces bundle adjustment to independent optimization subproblems that can be solved in parallel. We further apply Nesterov's acceleration and adaptive restart to improve convergence while maintaining its theoretical guarantees. Despite limited peer-to-peer communication, our method has provable convergence to first-order critical points under mild conditions. On extensive benchmarks with public datasets, our method converges much faster than decentralized baselines with similar memory usage and communication load. Compared to centralized baselines using a single device, our method, while being decentralized, yields more accurate solutions with significant speedups of up to 953.7x over Ceres and 174.6x over DeepLM. Code: https://joeaortiz.github.io/daba.