search

dotchen / LAV

(CVPR 2022) A minimalist, mapless, end-to-end self-driving stack for joint perception, prediction, planning and control.
https://dotchen.github.io/LAV/
Apache License 2.0
397 stars 68 forks source link

How can I visualize the results? #13

Closed kevinchiu19 closed 2 years ago

kevinchiu19 commented 2 years ago

Thanks to the author for the wonderful code! I would like to ask how to visualize the results of the network output, for example: RGB camera inputs, predicted road geometries, and detection and motion predictions respectively.

Looking forward to your reply, thank you very much!

dotchen commented 2 years ago

Thank you very much for your interest in our project.

You may use the visualization function here: https://github.com/dotchen/LAV/blob/a63af0999c2d696f5596060131a22087140044f2/lav/utils/logger.py#L94-L162

It overlays object detection, road segmentation and motion predictions in the visualization.

kevinchiu19 commented 2 years ago

Thank you for your reply！

Sorry I'm not very familiar with wandb, what is 'it' and 'opt_info'? Are there any examples to refer to?

kevinchiu19 commented 2 years ago

I just want to get the visualization like in the video (https://www.youtube.com/watch?v=-TlxbmSQ7rQ), the 'Predicted map' and 'Predicted traffic & planning'. Can I visualize it without using wandb?

dotchen commented 2 years ago

Yes, you should be able to run the above codes without wandb, just remove the wandb related lines and visualize the pyplot either in a default pop window or inline in a notebook.

Example usage of the function is in: https://github.com/dotchen/LAV/blob/main/lav/train_full.py#L20-L28

kevinchiu19 commented 2 years ago

Thank you for your patient reply. With this example, I can understand it immediately. Thank you very much!