dataset - 如何获取基蒂数据集中用于深度预测的密集地面真相图像？(How to obtain a dense ground truth image for depth prediction in kitti dataset?)

Question

For example let's say that this is the input image:

(例如，假设这是输入图像：)

Input image

(输入图像)

This is the original ground truth downloaded:

(这是下载的原始地面真理：)

GT

(GT)

But in all the papers they show something like this:

(但是在所有论文中它们都显示如下：)

heat map

(热图)

How can I obtain the last one?

(我如何获得最后一个？)

  ask by aca06 translate from so

Answer 1

In order to obtain a dense depth map, you need to run a depth inpainting/depth completion method on the Lidar data, which is the ground truth data you downloaded.

(为了获得密集的深度图，您需要对Lidar数据（下载的地面真实数据）运行深度修复/深度完成方法。)

I suggest you to take a look at the following:

(我建议您看看以下内容：)

https://github.com/balcilar/DenseDepthMap

(https://github.com/balcilar/DenseDepthMap)

https://gist.github.com/ialhashim/be6235489a9c43c6d240e8331836586a

(https://gist.github.com/ialhashim/be6235489a9c43c6d240e8331836586a)

Also the works under the KITTI Depth Completion Benchmark could be helpful:

(同样，根据KITTI深度完成基准进行的工作可能会有所帮助：)

http://www.cvlibs.net/datasets/kitti/eval_depth.php?benchmark=depth_completion

(http://www.cvlibs.net/datasets/kitti/eval_depth.php?benchmark=depth_completion)