Visual SLAM data set (3): KITTI data set

1. Official website

Official website address: https://www.cvlibs.net/datasets/kitti/eval_odometry.php

The Kitti data set is very large, including binocular, optical flow, scene flow, depth, odometry, target, tracking, road, semantics, original data and other major categories. Each major category contains many subdivided sequences. The category details are explained clearly on the official website.
Insert image description here

2. Register and log in

Click Register, enter your email and password, and then Register.

Insert image description here

Open your email and click the verification link to activate your account.
Insert image description here

There will be a prompt upon successful activation.

Insert image description here

Log in using your account and password.

3. Download the data set

For example, if I want to download the odometer data set of autonomous driving (grayscale image, 22GB), just click the download link. If your download speed is very slow, it is recommended to download directly from the Baidu network disk link.

4. Test data set

You can run the data set in a binocular program to obtain the camera trajectory and key frame trajectory, and then use the evaluation tool to evaluate various indicators.

Guess you like

Origin blog.csdn.net/qq_42257666/article/details/130540262