OpenVINO Overview

OpenVINO is the #7 ranked solution of our top AI Development Platforms. It's rated 4.0 out of 5 stars, and is most commonly compared to TensorFlow: OpenVINO vs TensorFlow

What is OpenVINO?

OpenVINO toolkit quickly deploys applications and solutions that emulate human vision. Based on Convolutional Neural Networks (CNNs), the toolkit extends computer vision (CV) workloads across Intel hardware, maximizing performance. The OpenVINO toolkit includes the Deep Learning Deployment Toolkit (DLDT).

OpenVINO Reviews

Filter by:
Filter Unavailable
Company Size
Filter Unavailable
Job Level
Filter Unavailable
Filter Unavailable
Filter Unavailable
Order by:
  • Date
  • Highest Rating
  • Lowest Rating
  • Review Length
Showingreviews based on the current filters. Reset all filters
Utku Aktas
Computer Vision Engineer at UpWork Freelancer
Real User
Nov 25, 2020
Fast, reduces the required processing power, and easy to manage

What is our primary use case?

I used OpenVINO for my deep learning solutions. I was working on some traffic management problems and had to detect pedestrians, cars, and some accidents via traffic videos. So, I used OpenVINO to implement my deep learning solutions to some servers. The OpenVINO software was on the local machine where we use the software. I used the OpenVINO hardware Mustang-V100-MX8, which includes eight Myriad Cores. I used OpenVINO software with these cards. We had two people focusing on OpenVINO. My colleague was dealing with communication about the project. He was communicating with Intel and other… more »

Pros and Cons

  • It reduces the required processing power on the CPU and GPU. With OpenVINO, you can run your normal algorithms and normal software on CPU, but don't require a huge amount of processor power. It is faster, and you have plenty of more resources for other jobs. It is easy to manage the software with OpenVINO. You can change the number of models or quotes. I can use five quotes for a model, or I can write a particular model with a quote. Management is easy without touching the software.
  • Their resolution time for certain kinds of issues could be better. I had a problem during the implementation, and it took them two or three months to resolve it. I wasted so much time. If it is a simple problem or implementation issue, they will provide the answers and solve it quickly, but if there are some problems with the product, it can take time.

Cost and Licensing Advice

  • Specifically for our products, it was costly to use OpenVINO without using the OpenVINO hardware. This was because we had not used the same number of CPUs. We had reduced the CPU number, but we didn't reduce it too much. Therefore, the cost was more with OpenVINO, but the efficiency was also much higher. We didn't have any problems with CPU usage or memory usage. OpenVINO really helped us with these issues.

What other advice do I have?

Reading the documents is really helpful, and there are multiple people who use the software and share their quotes, ideas, and documentation. It really helps in understanding how to implement. If there are any questions, you can also use Intel's forums. The community is growing, which is really helpful for implementation issues. By using this solution, I have learned to not change the version too frequently. When I was implementing OpenVINO, a new version come out, and I updated the version, but the code was incompatible. So, I had to change things. One or two months later, a new version came…