Following this article https://www.youtube.com/watch?v=fzYe_E5sARA&list=PLg-UKERBljNzXUIDjeb8oF-KRwp2fTU4i&index=38 , the assynchronous operation is described at the top of the image. At one time, there is only one inference that is executed.
However, to reduce the inference time, then I want to modify the asynchronous operation of NCS2 likes the graph at the below of the image. If I can do that I will save several inference time. Unfortunately, I can't find out any information about that.
Can you explain more for me about Asynchronous of NCS2? and Can I modify the asynchronous operation of NCS2 like I described before? If you have any information please let me know.
Thank you very much.
This might help you understand and achieve you target: https://techdecoded.intel.io/essentials/optimize-deep-learning-inference-applications-using-openvino-toolkit/#gs.jfaqz5
This also contain thorough details for optimization purposes: https://docs.openvinotoolkit.org/latest/openvino_docs_optimization_guide_dldt_optimization_guide.html
Intel will no longer monitor this thread since we have provided a solution. If you need any additional information from Intel, please submit a new question