search

David0tt / DeepGTAV

A system to easily extract ground truth training data for different machine learning tasks from GTAV
GNU General Public License v3.0
89 stars 9 forks source link

The format of DeepGTAV-Visdrone #16

Closed RoyAmoyal closed 10 months ago

RoyAmoyal commented 11 months ago

Are the classes the same as VisDrone?

And I noticed that after the class there are 4 numbers. Is it the same as the yolov5 format? Thanks!

David0tt commented 10 months ago

Yes, when using the VPilot script "datageneration_VisDrone.py" the generated classes are the same as VisDrone in the Format that is needed as input for YOLOv5.

In general, the classes returned in the message['bbox2d'] are specified very precisely. They are then generally translated/parsed into the format that fits the specific dataset, for example using the function parseBBoxesVisDroneStyle from from utils.BoundingBoxes.