Yiyi Liao

Hi, there are several packages you can use in Python, e.g., open3d, trimesh.

Thank you for your questions. We will release the dataloader of the baselines after cleaning up the code. For projecting 3D bounding boxes to velo_points, the 3D bounding boxes are...

Hi @Pyrestone , we released [accumuLaser](https://github.com/autonomousvision/kitti360Scripts/tree/master/kitti360scripts/devkits/accumuLaser) which we use to convert point clouds to the world coordinate and to accumulate them. You can use this to find the correspondence between...

Hi @leezonpen , we did not release any labels before frame 4300 of drive_0002 as they will be used as a part of our held-out test set. We have discarded...

Hi, I am not familiar with cv2.ominidir() function. Comparing the projectPoints() function of cv2.ominidir(): https://github.com/opencv/opencv_contrib/blob/master/modules/ccalib/src/omnidir.cpp#L143-L165 to our projection function: https://github.com/autonomousvision/kitti360Scripts/blob/master/kitti360scripts/helpers/project.py#L174-L198 I think `gamma1` and `gamma2` in our code correspond to...

Hi, sorry for the late reply. We plan to release our CRF code in another repo after some cleaning up. I will let you know when it is released.

Hi, sorry we do not have scripts to make rosbag of kitti360. The easiest way would be to structure KITTI-360 data following KITTI and use [kitti2bag](https://github.com/tomas789/kitti2bag). You may also adapt...

Hi, the dataset is collected on 28.05.2013. You can find latitude and longitude coordinates in our download page:

Hi @griffintin , thanks for your question. You can find instanceDist.json here:  The 2D instance segmentation masks can be considered as labels for panoptic segmentation, where every pixel is...

Hi @griffintin , > Thank you for your quick reply. I did not find .json file as you mentioned in "3D bounding box" data. Anyway, I made instance.json from .png...