search

ZikangZhou / HiVT

[CVPR 2022] HiVT: Hierarchical Vector Transformer for Multi-Agent Motion Prediction
https://openaccess.thecvf.com/content/CVPR2022/papers/Zhou_HiVT_Hierarchical_Vector_Transformer_for_Multi-Agent_Motion_Prediction_CVPR_2022_paper.pdf
Apache License 2.0
577 stars 115 forks source link

Visualization tool for local region #9

Closed Jihong-Kim97 closed 2 years ago

Jihong-Kim97 commented 2 years ago

Hello, First of all, thanks for your excellent work! I wonder if there is any visualization code or tool for local regions. I know that there is visualization tool for scenario which is provided in argoverse-api but I want to watch just local region that local encoder use to train or infer. Thank you :)

ZikangZhou commented 2 years ago

Hello,

I didn't clean that part of the code, but my visualization tool is modified from the baseline repo. In general, you can use get_lane_ids_in_xy_bbox() in the map API with the target agent's position@t=20 as the query center to get nearby lanes. Some segments of these nearby lanes may not be used in training, because some lanes may be pretty long and I filter out those segments whose distance to the target agent is larger than the pre-defined radius.