search

tsinghua-rll / VoxelNet-tensorflow

A 3D object detection system for autonomous driving.
MIT License
453 stars 123 forks source link

Is it possible to train for pedestrian and cyclists? #49

Open Abhirami20 opened 6 years ago

Abhirami20 commented 6 years ago

Can this code be used for training all the three class Car, Pedestrian and cyclists together? In code, it's for car only and can be trained for individual classes. Thanks in advance.

Abhirami20 commented 6 years ago

@jeasinema Could you please guide me regarding my doubt?

Abhirami20 commented 6 years ago

@LinHungShi Could you please see this issue?

LinHungShi commented 6 years ago

No, you can't train all the classes without modifying the code.

Abhirami20 commented 6 years ago

@LinHungShi Does preprocessing to all

No, you can't train all the classes without modifying the code. Thank you for your reply!

Does preprocessing effect training on all classes? Now I am working on it but I understood that voxelnet- size required is different for pedestrian/ cyclist and car. So I wants to know if just rewriting code is enough or not?

LinHungShi commented 6 years ago

The parameters for car and pedestrian/cyclist are a bit different in the paper. However you can assume they are the same.

Abhirami20 commented 6 years ago

The parameters for car and pedestrian/cyclist are a bit different in the paper. However you can assume they are the same.

Thank you so much for the guidance! :) Is editting config.py enough for training with all classes at once or do I need to edit any other codes?