X-SLAM: Scalable Dense SLAM for Task-aware Optimization using CSFD

ACM Transactions on Graphics (SIGGRAPH 2024)

Zhexi Peng1, Yin Yang2, Tianjia Shao1, Chenfanfu Jiang3, Kun Zhou1
1Zhejiang University, 2University of Utah, 3University of California, Los Angeles
MY ALT TEXT

Real-time robot active scanning and reconstruction with semantic segmentation based on our X-SLAM system. With X-SLAM, robots can carry out automatic navigation and scanning within an unknown environment (left), and obtain a reconstruction with semantic segmentation (middle). The scanning process is presented on the right. We propose the first real-time differentiable dense SLAM system utilizing CSFD. By integrating it with a neural network, we facilitate robot active scanning and scene comprehension with semantic awareness.

Abstract

We present X-SLAM, a real-time dense differentiable SLAM system that leverages the complex-step finite difference (CSFD) method for efficient calculation of numerical derivatives, bypassing the need for a large-scale computational graph. The key to our approach is treating the SLAM process as a differentiable function, enabling the calculation of the derivatives of important SLAM parameters through Taylor series expansion within the complex domain. Our system allows for the real-time calculation of not just the gradient, but also higher-order differentiation. This facilitates the use of high-order optimizers to achieve better accuracy and faster convergence. Building on X-SLAM, we implemented end-to-end optimization frameworks for two important tasks: camera relocalization in wide outdoor scenes and active robotic scanning in complex indoor environments. Comprehensive evaluations on public benchmarks and intricate real scenes underscore the improvements in the accuracy of camera relocalization and the efficiency of robotic navigation achieved through our task-aware optimization. The code and data are at https://gapszju.github.io/X-SLAM

Overview

Robot activate scanning and reconstruction

Camera relocalization

More results

Comparison