OpenVino Reviews & Product Details

The most important upside of OpenVino is ability to predict real time on CPU. It also has various accelerators like GPU, VPU, FPGA. OpenVino's documentation is very well maintained. Hence, its easy to use. We can also customize OpenCL. We can also prune and quantize deep learning models. It has its own benchmarking tool. It has many model conversion features. Like converting any model to its intermediate representation from onnx,pytorch,tensorflow, keras.

It has many sample Deep learning/computer vision examples that are already well optimized. Review collected by and hosted on G2.com.

It has many versions. So you need to stay updated in our to run various DL models efficiently. You might get version conflicts. Its feature of model optimization is a bit slow. It becomes difficult to convert latest state of the art models due to internal layer implementation. Training complex neural network can be a concern as model conversion can be quite typical. Also it does not have more references for beginners. These are the things I do not like about Openvino and needs improvements. Review collected by and hosted on G2.com.

I majorly like that OpenVino available for most of the platforms, which is one of my top features. Apart from this, the workflow and the documentation of how to use it are unique too. I majorly like that OpenVino available for most of the platforms, which is one of my top features. Apart from this, the workflow and the documentation of how to use it are unique too. I advise those who are seeking to learn something latest up-to-the-date toolkit I must suggest using this and as per my experience, they will know and built a lot of new things. Review collected by and hosted on G2.com.

When I was commencing this, I found little difficulty, but after some time, I understood all very well. Review collected by and hosted on G2.com.

OpenVino has a lot of resources and examples, which can be utilized to build our own custom deep learning solutions. OpenVino has worked very well in industrial projects, especially in custom computer vision solutions. Also, it is very easy to integrate with Python projects and can run on a plethora of devices, which makes it an ideal choice as an inference engine. Review collected by and hosted on G2.com.

The learning curve is a bit steep for beginner and novice developers. Review collected by and hosted on G2.com.

Increased inference speed - My inference time reduced from 120ms to 30ms using OpenVINO Review collected by and hosted on G2.com.

Little complex code to save and use loaded model Review collected by and hosted on G2.com.

Deep Learning models support, IR Format, Model Optimization, Ngraph Bridge, etc. The list goes on... Review collected by and hosted on G2.com.

Some Models failed by model optimizers to convert IR Files. And the conversion should be both ways optimized and not optimized; sometimes, I want to retain my ops and nodes. Review collected by and hosted on G2.com.

I like the pre-trained models. I like the ease of setting up and getting started. Review collected by and hosted on G2.com.

With each iteration they change things and the file structure or hierarchy so certain things needed to be changed. Also with different iterations there was a need to change plugins. Review collected by and hosted on G2.com.

Using OpenVino to perform inference on the Intel integrated GPU helped my team lower CPU utilization while making use of previously unutilized resources. Review collected by and hosted on G2.com.

There is a lack of support for kernel tuning and optimization granularities with the integrated GPU inference Review collected by and hosted on G2.com.

*user friendly

*Easy to use

*Deep learning models are already available Review collected by and hosted on G2.com.

*It only provides inferences

*Pre trained models are required Review collected by and hosted on G2.com.

Optimize and deploy deep learning solution across multiple Intel platforms.

Allows to maximize the performance on any type of processors. Review collected by and hosted on G2.com.

Sometimes Openvino don't able to convert deeplearning models.

To resolve this you need to create your own custom layers and it's difficult. Review collected by and hosted on G2.com.

I guess the best part is that it works once you spend a lot of time learning it. Review collected by and hosted on G2.com.

Not straightforward to learn and the documentation is sparse. The documentation that does exist is not well organized. Review collected by and hosted on G2.com.