本文调研了目前比较先进的基于传统方法和基于学习方法的视觉SLAM系统，

This paper investigates the current relatively advanced visual SLAM systems based on traditional methods and learning methods, and analyzes the advantages and disadvantages of each system. The model and structure of the convolutional neural network, as well as the application in the visual SLAM front-end are studied, which lays a theoretical foundation for the realization of feature extraction using the convolutional neural network.

This paper investigates the current advanced visual SLAM system based on traditional methods and learning methods, and analyzes the advantages and disadvantages of each system. The model and structure of reel neural network and the application of visual SLAM front end are studied, and the theoretical basis is laid for the characteristic extraction using reel neural network.

This paper investigates the advanced visual slam system based on traditional methods and learning methods, and analyzes the advantages and disadvantages of each system. The model, structure and application of convolutional neural network in the front end of visual slam are studied. It lays a theoretical foundation for feature extraction by convolution neural network.<br>