search

MediaBrain-SJTU / CoCa3D

Other
78 stars 3 forks source link

How does OPV2V+ get? #2

Open chaoyiwang09 opened 1 year ago

chaoyiwang09 commented 1 year ago

Hi,

  1. In your work, you used OPV2V+ instead of OPV2V by including more collaborative agents. Could you instruct me how you include more collaborative agents?

  1. The hypes_yaml file folders include many config files. Which one is the camera-only 3d detection method mentioned in this paper?

  2. There's something wrong downloading DAIR-V2X and is CoPerception-UAVs+ an open dataset?

    1. In your yaml of opv2v, you have valid_data: "/GPFS/rhome/yifanlu/OpenCOOD/dataset/OPV2V_MoreAgents/dataset_annos/valid_data_info.pkl". According to Data Preparation, The annotation is in the yaml files rather than the valid_data_info.pkl. How could I download the valid_data_info.pkl?
sidiangongyuan commented 1 year ago

Have you solved this question now? I have the same question. I just found that OPV2V does not include depth information. So I am curious about how this work did?