search

SPengLiang / DID-M3D

[ECCV 2022] DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection.
MIT License
76 stars 11 forks source link

Visualization #23

Closed Xiaolong-RRL closed 1 year ago

Xiaolong-RRL commented 1 year ago

Thanks for you nice work!!

But I'm still confused about how you visualize this figure (Fig.5 in paper) ?

image

SPengLiang commented 1 year ago

Sorry for the late reply! We use mayavi to visualize the point cloud (only used in visualization) and 3D detection results. You can refer to this codebase (https://github.com/kuixu/kitti_object_vis) for visualization.

Xiaolong-RRL commented 1 year ago

I got it! Thanks!!

SPengLiang commented 1 year ago

You can also use open3d for visualization. You can refer to this codebase https://github.com/chaomath/open3d-kitti-visualization. Both mavayi and open3d can generate similar results.

Xiaolong-RRL commented 1 year ago

OK~ I will try both, thank you very much for your enthusiastic reply!!