hkust-vgd / lcd

[AAAI'20] LCD: Learned Cross-Domain Descriptors for 2D-3D Matching
BSD 3-Clause "New" or "Revised" License
134 stars 19 forks source link

About test the registration of 2D image and 3D point cloud data #6

Open houyongkuo opened 3 years ago

houyongkuo commented 3 years ago

Hi!Thanks for your insteresting works. If I want to test the registration of 2D image and 3D point cloud data, is there any requirement for the format of the 2D image? If using .jpg format, can I directly run “python -m apps.align_point_cloud samples/.jpg samples/.ply --logdir logs/LCD-D256/”?

Hope to show a demo about producing a 2D-3D registration

Thanks. Best wishes!

Jason-Yxj commented 2 years ago

Hi!Thanks for your insteresting works. If I want to test the registration of 2D image and 3D point cloud data, is there any requirement for the format of the 2D image? If using .jpg format, can I directly run “python -m apps.align_pointcloud samples/.jpg samples/_.ply --logdir logs/LCD-D256/”?

Hope to show a demo about producing a 2D-3D registration

Thanks. Best wishes!

Hi!Have you tried testing the registration of 2D image and 3D point cloud data? The file "app.align_point_cloud" can't test it.

houyongkuo commented 2 years ago

嗨!感谢您有趣的作品。如果我想测试2D图像和3D点云数据的配准,对2D图像的格式有要求吗?如果使用.jpg格式,可以直接运行“python -m apps.align_pointcloud samples/ .jpg samples/_ .ply --logdir logs/LCD-D256/”吗? 希望展示一个关于生成 2D-3D 配准的演示 谢谢。最良好的祝愿!

您好!您尝试过测试 2D 图像和 3D 点云数据的配准吗?文件“app.align_point_cloud”无法测试。

not yet