Katsuya Hyodo

> Have you tried to run it on the MyriadX or on the Intel GPU via Openvino (the original converted model was working only on the CPU) ? I see,...

Hmmm. I have tested it by specifying GPU as the device, but it seems oddly slow. In my environment, I doubt that the GPU is really being used. - Model...

I reinstalled OpenCL and it worked on GPU. :+1: https://user-images.githubusercontent.com/33194443/190673680-74c87f90-6cc6-4cc7-99bf-cbbe9288a450.mp4

Perhaps the normalization of the model's entrance may have influenced it. I will try some patterns too. ``` # -1 to 1 norm x = x / 127.5 - 1.0...

When **`MYRIAD`** was specified, nothing was detected. There appears to be a problem with the internal logic of NCS2 (MYRIAD). - Model 256x320 - 10 persons - Enable Normalization [openvino_256x320_.zip](https://github.com/geaxgx/depthai_movenet/files/9591056/openvino_256x320_.zip)...

I agree. Because of that past history, I have not conducted Myriad tests for several more years. :smile_cat:

Thank you. I added a task to TODO. I do not know when it will be, but I will do my best. **https://github.com/PINTO0309/OpenVINO-DeeplabV3/wiki/TODO**

Explain the meaning of my words. - OpenVINO officially supports armv7l(ARM) - My test program does not support the combination of NCS2 + ARM(RaspberryPi / TX2) - Argmax and some...

When using OpenVINO, benchmark results are faster with inference speed using a Intel's CPU than using NCS2. You will have no major benefit in purchasing NCS2. **[LattePanda Alpha + OpenVINO...

It has been confirmed to work with Celeron. However, because it does not support the latest Intel architecture, it seems that sufficient performance can not be achieved. OpenVINO is designed...