Added the 3D (X,Y,Z) position to what is handled by the xyz algorithm

ZFTurbo / Weighted-Boxes-Fusion

Set of methods to ensemble boxes from different object detection models, including implementation of "Weighted boxes fusion (WBF)" method.

MIT License

1.7k stars 237 forks source link

Closed RemiFabre closed 1 year ago

RemiFabre commented 1 year ago

Tested through the simple unit test and also runing : python3 luxonis_yolo_helper.py from : https://github.com/pollen-robotics/simple_grasp_pipe/tree/master/simple_grasp_pipe/utils => Meaning I had the camera in hand and tested the change on several items and setups, the results seem coherent

ZFTurbo commented 1 year ago

Can you please explain why we need (x, y, z) positions? I think it should be added as spearate algorithm. Definetly not in the main function.

RemiFabre commented 1 year ago

My apologies, this wansn't meant to be a PR here, but in the fork I've made. I'll close this.