Show HN: Object Detection in an Hour

Re:Show HN: Object Detection in an Hour by notum on Saturday August 07, 2021 @04:43PM (#28101655)

Looks great! Very innovative approach. Are the generated models compatible with OpenCV OAK camera?

[ Parent | on HN ]

Re:Show HN: Object Detection in an Hour by wokwokwok on Saturday August 07, 2021 @10:32PM (#28103776)

https://github.com/facebookresearch/detectron2/issues/1602

^ probably not, since they use detectron2, but given the labeled images are really the core part of this, there’s no reason you can’t use them on a different mode that is compatible.

[ Parent | on HN ]

Re:Show HN: Object Detection in an Hour by hietalajulius on Sunday August 08, 2021 @11:05AM (#28107248)

Yeah the labels are loaded into the Detectron2 format from the 3D annotations json at train time, we plan to add similar data loading for YOLOv3 etc soon. Starting out with Detectron2 was mainly for POC/demo purposes, the idea in the future is to be able to feed the data anywhere it might be needed.

[ Parent | on HN ]

Re:Show HN: Object Detection in an Hour by hietalajulius on Sunday August 08, 2021 @10:50AM (#28107128)

I don't know the specifics of that camera/its software, but the trained models are saved as TorchScript (https://pytorch.org/docs/stable/jit.html) which can be used very flexibly in python/C++.

[ Parent | on HN ]