We just raised a $30M Series A: Read our story

OpenVINO OverviewUNIXBusinessApplication

OpenVINO is #3 ranked solution in top AI Development Platforms. IT Central Station users give OpenVINO an average rating of 8 out of 10. OpenVINO is most commonly compared to TensorFlow:OpenVINO vs TensorFlow. The top industry researching this solution are professionals from a manufacturing company, accounting for 30% of all views.
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 Buyer's Guide

Download the OpenVINO Buyer's Guide including reviews and more. Updated: November 2021

Pricing Advice

What users are saying about OpenVINO pricing:
  • "We didn't have to pay for any licensing with Intel OpenVINO. Everything is available on their site and easily downloadable for free."

OpenVINO Reviews

Filter by:
Filter Reviews
Industry
Loading...
Filter Unavailable
Company Size
Loading...
Filter Unavailable
Job Level
Loading...
Filter Unavailable
Rating
Loading...
Filter Unavailable
Considered
Loading...
Filter Unavailable
Order by:
Loading...
  • Date
  • Highest Rating
  • Lowest Rating
  • Review Length
Search:
Showingreviews based on the current filters. Reset all filters
MM
Systems and Solutions Architect at a tech services company with 1,001-5,000 employees
Real User
Top 20
Open-source, easy to integrate, and perfectly tailored to the Movidius chipset

Pros and Cons

  • "The initial setup is quite simple."
  • "At this point, the product could probably just use a greater integration with more machine learning model tools."

What is our primary use case?

We currently make technology that uses the Intel VPU, the Movidius chipset. We run OpenVINO on it.

It's for Edge IoT. We make the hardware and we cater to customers who are looking for Edge IoT solutions, and the product is really for edge processing or video co-processing for machine vision. We distribute that data on the customer's network using our Edge solution, which is based on DDS, distributed data system. Basically, we use it for machine vision applications.

What is most valuable?

The solution's ability to stream data directly from camera inputs is the most valuable aspect for us. 

It's tailored to the Movidius chipset, which makes it a nice package. You don't have to run it on the Movidius. It runs on X86, however, we'd like to use it with our Movidius based co-processor. 

The ease of integration is fantastic. The option to run it just on X86 or X86 plus an Intel CPU is great.

It's an open-source solution.

The initial setup is quite simple.

What needs improvement?

Generally, when you deploy edge products, it's really about latency. It's about getting that camera input, being able to process it, extracting the information you need, and getting the solution back to the person who made the request. Although I'm not necessarily saying its latency or accuracy is bad, it's always something that can be improved upon. By focusing on improving these areas, they can make the overall solution even better.

At this point, the product could probably just use a greater integration with more machine learning model tools. However, that's not advice from experience per se. That's always just helpful in general. To be able to incorporate more models into the product makes it stronger. Therefore, to be clear, it's not coming from a point of a current deficiency. It's just a general comment.

For how long have I used the solution?

We've used the solution for a while now. We've used it over the course of the last 12 months as well. My personal experience has been a bit more limited and I would say that I've been using it for the last eight or so months.

What do I think about the stability of the solution?

The solution is quite stable. There are no bugs or glitches. It doesn't crash or freeze. It's reliable.

What do I think about the scalability of the solution?

The scalability seems to be pretty good. We've had good feedback on it. Our engineers seem to like it, and, for me as a kind of an end-user, it seems to be working fine.

How are customer service and technical support?

As gold partners with Intel, we do get pretty good support. I do get feedback from them. I have key contacts that I can reach out to directly, and then they're fairly responsive. We've been quite satisfied so far.

How was the initial setup?

The initial setup, for me at least, was pretty easy and straightforward. It is integrated with our binary image for our platform, so it already comes with it. The engineers already integrated it with our hardware solution and it comes to me as a binary that I install. It makes everything very simple. I wouldn't describe the setup as complex.

What was our ROI?

That's going to be really on our customers. We're not using it as a revenue generation tool. We use it as a development enhancement and then we sell our solutions to the end customer. The ROI is really a question for the end-user, on how they can realize their initial investment back.

What's my experience with pricing, setup cost, and licensing?

This solution is made available as an open-source product.

Which other solutions did I evaluate?

I have done a cursory bit of research into the product against a few other options.

What other advice do I have?

I'm not exactly sure which version of the solution we're using. I'm assuming it's fairly current as we're deploying our Edge IoT platform using it.

While the solution is deployed on-premise, but we have the availability to hook into Amazon web services and Microsoft Azure.

OpenVINO's part of Intel's framework and we're a gold partner with Intel.

I would say it's a very good vision processing unit. It's a very good VPU if you're using Intel type of architectures as a co-processor. I have only really had experience with Intel with any OpenVINO based on the Movidius and therefore I'd like to get more hands-on time with others. However, generally, it's a good platform. It's worth exploring and can handle multiple camera streams, and it's straightforward to use. A company would benefit from trying it out.

In general, I would rate the solution at an eight out of ten.

Which deployment model are you using for this solution?

On-premises
Disclosure: I am a real user, and this review is based on my own experience and opinions.
Flag as inappropriate
ChristianRivera
Freelance Engineer at Autonomo
Real User
Top 5Leaderboard
Has good model comparison, model testing, evaluation, and deployment features

Pros and Cons

  • "The features for model comparison, the feature for model testing, evaluation, and deployment are very nice. It can work almost with all the models."
  • "It has some disadvantages because when you're working with very complex models, neural networks if OpenVINO cannot convert them automatically and you have to do a custom layer and later add it to the model. It is difficult."

What is our primary use case?

OpenVINO is good for budgets because you don't have a computer vision model for classification for object detection obligations. You can run it on a server with Azure but it can be costly. Sometimes the application has to be on heavy dedicated hardware, like a small computer. In this case, machine learning applications are not so good because they demand a lot of computer resources and a lot of CPU resources are not so fast. In terms of accuracy and speed trade-off performance, you have to sacrifice a bit of accuracy in your inference in order to get better speed. When you deploy models with the OpenVINO format into devices like PC boards, it's a great tool. The online testing platform that Intel has for OpenVINO is really nice.

It's sort of like a sandbox environment. You want to test what kind of hardware is available. You can test it and watch what works better in talking about the preference. Later you can decide based on the budget.

What is most valuable?

The features for model comparison, the feature for model testing, evaluation, and deployment are very nice. It can work almost with all the models. 

What needs improvement?

It has some disadvantages because when you're working with very complex models, neural networks, if OpenVINO cannot convert them automatically and you have to do a custom layer and later add it to the model. It is difficult. These are the main disadvantages that OpenVINO has that are a bit limited for some models.

For how long have I used the solution?

I have been using OpenVINO for two years. 

What do I think about the stability of the solution?

Stability is good. It's growing fast and they implement fixes for old issues. I hope the installation and usage could be more clear and easier to implement for the deployment. 

What do I think about the scalability of the solution?

Scalability is not so straightforward, it's not so easy. When working with some applications, using some of our scaling options to manage those multiple containers is not so easy. But it is a good tool, it's a good tool for working with Intel devices.

How was the initial setup?

It took some time to install. Using the Docker container is not very straightforward. It raises many errors. The main installation has a lot of steps and there are many details that are easy to miss. You have to do it two or three times to get it right. It could take six to eight hours to set up. You have to test and build.

What's my experience with pricing, setup cost, and licensing?

I have been using the open-source version, so it's free. 

What other advice do I have?

Install the latest version that already has fixes for old problems. Work with some neural network, with a few layers to test it. If you use a neural network, like a fast R-CNN, it wouldn't work because it's too complex and there are some layers that are not recognized by OpenVINO. Start small and continue growing. Make an account in the Intel OpenVINO platform so you can start testing with Jupyter Notebook and send your inference jobs to different kinds of devices. Check the performance, know the differences between the different hardware, and how you can site the project. It is a good platform. The price I have seen is not so expensive. That's my advice.

OpenVINO is a good tool if you want to work not in the cloud, you have to work on the edge. If you want to adapt your models, it has a good pipeline to follow. You have to learn it well because it could take time to debug some of the errors. Some of them are not very explained and you have to go through your codes a bit blind.

I would rate it a nine out of ten because it's a very good tool. It's not complete but is the best in the market right now.

Disclosure: I am a real user, and this review is based on my own experience and opinions.
Learn what your peers think about OpenVINO. Get advice and tips from experienced pros sharing their opinions. Updated: November 2021.
552,305 professionals have used our research since 2012.
Zafar Muhammed
Machine Learning Software Developer at Freelancer
Real User
Top 5Leaderboard
A free toolkit providing improved neural network performance

Pros and Cons

  • "The inferencing and processing capabilities are quite beneficial for our requirements."
  • "The model optimization is a little bit slow — it could be improved."

What is our primary use case?

I created a retail recognition custom model — a model on RPX 2017. Afterward, I transferred it to OpenVINO for object detection and retail detection.

We have a team of three people who use OpenVINO.

How has it helped my organization?

It's a great solution for getting camera images, processing them, and extracting the reserves. It's better and more cost-effective than using Intel myriad X.

What is most valuable?

The inferencing and processing capabilities are quite beneficial for our requirements.

Compared to Jetson Nano or Jetson TX2, or Jetson Xavier, OpenVINO is a much more cost-effective solution. Processing-wise, they are comparable to Jetson and maybe Jetson Xavier NX.

What needs improvement?

The model optimization is a little bit slow — it could be improved. They should introduce some type of deep learning accelerator, like Jetson Xavier NX.

There is a lacking in vehicle recognition — types of vehicles. Differentiating between cars, SUVs and different types of light, heavy, and medium trucks can be tricky. We have to train such models ourselves and then transfer them onto OpenVINO.

For how long have I used the solution?

I have been using this solution for roughly 18 months.

What do I think about the stability of the solution?

It works quite well, but regarding the frames per second — If we want to add on a few more cameras to the hardware, we can't do that.

By assigning different IPs to our own software, we can access different cameras on a real-time basis, to analyze the traffic from different lanes of vehicles and roads.

This solution has never crashed and we haven't experienced any bugs; however, there are improvements that need to be made.

What do I think about the scalability of the solution?

I didn't scale it, but I want to build more than one board, on top of each other, so that we can improve the processing compute capabilities.

How are customer service and technical support?

The technical support is very good.

How was the initial setup?

I didn't experience any difficulties when setting up this solution.

What about the implementation team?

We have a really small team of about two to three people. One person has to go to the site and set up the hardware so we can access the board remotely. We can work from our office or any location. It's a one-time installation process; from there, our team can work from the office.

We were planning on deploying this solution on different devices, so the planning aspect took a little bit of time, but overall, we were up and running within a week.

What's my experience with pricing, setup cost, and licensing?

We didn't have to pay anything for Intel OpenVINO, everything was available on their site. All of their solutions, inference engines, and other model optimizations are all available for free. 

We didn't have to pay for any licensing with Intel OpenVINO. Everything is available on their site and easily downloadable for free.

Which other solutions did I evaluate?

We are currently trying Nvidia and Xavier NX because some of their models and capabilities look promising.

We are comparing the processing and usability of these two boards and we are planning to have a model GPU board with us also.

What other advice do I have?

I would absolutely recommend OpenVINO. I think it's a good introduction to machine learning and inferencing.

On a scale from one to ten, I would give OpenVINO a rating of nine.

Which deployment model are you using for this solution?

On-premises
Disclosure: I am a real user, and this review is based on my own experience and opinions.
Product Categories
AI Development Platforms
Buyer's Guide
Download our free OpenVINO Report and get advice and tips from experienced pros sharing their opinions.