Deploy custom deep learning models on AD-3DSMARTCAM1-PRZ

Hi!
Thanks for the question. In case you can manage your further development and project without Ubuntu GUI interface, there is a specific reduced image that can be installed on the camera. This allows you to obtain more free space by deleting all the unnecessary packages. (Be aware that in case you install it, there is no image to install back the GUI, so far)

You can find the steps in the README for the Image setup, and also the necessary steps to install Cuda API on the following branch of the repo: https://github.com/analogdevicesinc/aditof_sdk/tree/add_cuda_utils
and the description: https://github.com/analogdevicesinc/aditof_sdk/tree/add_cuda_utils/bindings/cuda_utils

(This branch supposed to be a binding part for the SDK that processes the frames on GPU, however it is still under review, but feel free to use for your project.)

Best regards,

Robert

Hi,

Thanks for your reply! But I'm still wondering how AD-3DSMARTCAM1-PRZ is supposed to be used for deploying custom deep learning model? Because managing the development without Ubuntu GUI interface is not so preferable (can't use OpenCV to do some visualization).   

In case of an existing deep learning model, it can be implemented either in tensor flow or in other frameworks that support CUDA API. This can be implemented on the camera in order to do the processing. There is already an example on a branch of the repository for this implementation. But for development (especially with graphical aids) a more performant device is advised to be used.

In case of an existing deep learning model, it can be implemented either in tensor flow or in other frameworks that support CUDA API. This can be implemented on the camera in order to do the processing. There is already an example on a branch of the repository for this implementation. But for development (especially with graphical aids) a more performant device is advised to be used.