super-gradients
super-gradients copied to clipboard
YOLO-NAS + BOXMOT (YOLO-NAS multi-object tracking)
🚀 Feature Request
I have an example in my repository on how to integrate SOTA tracking methods together with Yolo-NAS.
I am using the Yolov8 backend functionality to achieve this. This is only because I am more familiar with it. I am certain it would be equally easy with the super-gradients backend :smile:
My pip package contain SOTA tracking methods: both motion only and motion + ReID. Pairing up boxmot together with yolo-nas is seamless and leads to very good results out of the box.
Proposed Solution (Optional)
Adapt this to super-gradients backend
I fixed some bugs. It is stable now.
Wow!! This is great! Thank you @mikel-brostrom for this fantastic contribution. We can't include the code in our repo, since unfortunately, it is under AGPL 3.0 license (our repo is Apache 2.0). But we will definitely add a Colab notebook on how to use BOXMOT + YOLO-NAS and points to your repo of course. Thanks :)
We can't include the code in our repo, since unfortunately, it is under AGPL 3.0 license (our repo is Apache 2.0)
Yes, I see
But we will definitely add a Colab notebook on how to use BOXMOT + YOLO-NAS
Nice! I will keep this open in case you need support integrating this :smile: