Can you specify the command used to run the demo and the models used along with?
I have tried the demo on latest OpenVINO version(2020.1) using the driver-action-recognition-adas-0002-encoder, driver-action-recognition-adas-0002-decoder models and it is working fine.
Thank you for your reply.
See the attached action_recognition_screenshot1.png for the command specification.
I specified action-recognition-0001-encoder + action-recognition-0001-decoder. (FP16)
The error is OK in the encoder specification, but occurs in the decoder specification. (Line 102 of the screenshot screen)
Maybe I think it's the latest, can you tell me how to know the version of OpenVINO?
Navigate to the OpenVINO install directory(default install directory is C:\Program Files (x86)\IntelSWTools)
If the latest version is installed, there should be a folder named openvino_2020.1.033. If you are using any older version of OpenVINO, please upgrade to the latest version and give a try.
Hope it solves the issue.