Hi,
Is it possible to run YOLOv2 on Drive PX2 and if so, how ?
How did YOLO based DriveNet run on Drive PX2?
Looks like its possible to run YOLOv1 on drive px2 using converted caffe model.
But in order to run YOLOv2, darknet need to be ported on Drive works.
Or Without Drive Works, something like docker is needed on Drive Px2.
And It seems like Someone did run docker based YOLO on Jetson board.
Best,
Dear kakpple,
We had released “DRIVE PX2 Deep Learning App Notes” on https://developer.nvidia.com/nvidia-drive-downloads.
Could you please refer to the application note for you topic? Thanks.
Hi Steve,
Is there darknet container for drive PX2 docker to run YOLO?
Jetson TX2 docker container for darknet is already available.
https://github.com/open-horizon/cogwerx-jetson-tx2/wiki/Yolo-and-Darknet-on-the-TX2
Can we run the above darknet docker container in Drive PX2 ?
Or else we need to cross compile darknet for Drive PX2
Or how do we run YOLO on PX2?
Thanks,
Hi kakpple,
I have tried yolo v2 on drivepx, it did not give good framerate at only around 20 hz per second. cuDNN suppose to speed it up but does not seem to work on drivepx2, perhaps different architecture. Anyway, unless you found a way to integrate gmsl camera frame to Mat in opencv, you left with only one usb port for your usb camera i suppose. Other solutions include ip camera. If you need a reference, can clone from here: GitHub - khaixcore/neural_cam_ros: This package is essentially a ros-wrapper of neural_cam. More features would be added in the future, geared towards mobile robot platform. Eventual goal of this package is to solve the problem of given an image, where is the obstacle with respect to the robot in a 3D space.
regards
haixun