UPGRADE YOUR BROWSER

We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

cancel
Showing results for 
Search instead for 
Did you mean: 
Observer marie-laurie
Observer
537 Views
Registered: ‎05-16-2018

About the possible outputs of CHaiDNN

Hi all, 

 

From the API document, only 3 types of output are supported by CHaiDNN (segmentation, classification and detection).

Is it possible to have different ouput? I mean is there a way for me to modify a part in the code ? 

Ideally, I would like to have as an output a high resolution image coming from a low resolution image.

Thanks in advance, 

 

ML

0 Kudos
2 Replies
Xilinx Employee
Xilinx Employee
510 Views
Registered: ‎07-25-2018

Re: About the possible outputs of CHaiDNN

Hi,

It is possible to modify the part of code which writes the data out. Please go through CHaiDNN-v2 documentation and let me know if you have more questions.
Explorer
Explorer
474 Views
Registered: ‎10-24-2008

Re: About the possible outputs of CHaiDNN

@marie-laurie@vkjainI understand that there are definitely applications, GANs and Super Resolution perhaps being two obvious examples, where there is a need to generate higher resolution images using CNNs.  I would personally be very interested (as I believe would the CHaiDNN team) to better understand your requirements if you are willing to share additional details.  If there is a general or more wide-spread need to employ FPGA inference of such topologies, perhaps this is something that the team can explore further.

 

My 2 cents....and worth every penny.

 

Cheers,

Quenton

0 Kudos