Graphite: A GPU-Accelerated Mixed-Precision Graph Optimization Framework
This enables faster large-scale optimization for robotics and computer vision applications, though it is incremental as it builds on existing graph optimization techniques.
The authors tackled the problem of large-scale graph optimization for applications like SLAM by developing Graphite, a GPU-accelerated mixed-precision framework. The result was up to 59x speed-up compared to CPU baselines while using less memory and maintaining generality.
We present Graphite, a GPU-accelerated nonlinear least squares graph optimization framework. It provides a CUDA C++ interface to enable the sharing of code between a real-time application, such as a SLAM system, and its optimization tasks. The framework supports techniques to reduce memory usage, including in-place optimization, support for multiple floating point types and mixed-precision modes, and dynamically computed Jacobians. We evaluate Graphite on well-known bundle adjustment problems and find that it achieves similar performance to MegBA, a solver specialized for bundle adjustment, while maintaining generality and using less memory. We also apply Graphite to global visual-inertial bundle adjustment on maps generated from stereo-inertial SLAM datasets, and observe speed-ups of up to 59x compared to a CPU baseline. Our results indicate that our framework enables faster large-scale optimization on both desktop and resource-constrained devices.