After designing and training, the model file needs to be converted into a ".dlc" (Deep Learning Container) file to be used by the Snapdragon NPE runtime.
I have a project on behavior recognition based on 3D convolution, and I would like to ask whether SNPE also supports 3D convolution, that is, whether models based on 3D convolution can also be converted into ".dlc " files to be used by the Snapdragon NPE runtime.
Opinions expressed in the content posted here are the personal opinions of the original authors, and do not necessarily reflect those of Qualcomm Incorporated or its subsidiaries (“Qualcomm”). The content is provided for informational purposes only and is not meant to be an endorsement or representation by Qualcomm or any other party. This site may also provide links or references to non-Qualcomm sites and resources. Qualcomm makes no representations, warranties, or other commitments whatsoever about any non-Qualcomm sites or third-party resources that may be referenced, accessible from, or linked to this site.
Hi,
Snapdragon NPE doesn't support 3D convolutions.
It is known that 3D convolution requires 4D input which is not supported by in NPE.