• Aerialtronics Logo
  • DroneVolt Logo


What makes the Pensar Camera + SDK unique?

The Pensar Camera + SDK combo is the future of AI in computer vision and includes:

  • 1 Pensar Camera with EO x30 zoom and IR sensors.
  • 1 Nvidia Jetson TX2 GPU
  • External connectivity (Micro HDMI, SD Card, Micro USB B)
  • Hardware-accelerated Computer Vision and AI Libraries
  • API for communication and hardware control
  • Ready-to-use, pre-trained neural networks or easily plug your own in
  • Capability to import neural networks in most common formats (caffe, darknet and soon tensorrt)
  • A complete, plug-in based, application framework
  • Development toolchain

What is included in the different packages?

All the prices include:

Pensar Camera
  • Visible x30 zoom and Thermal cameras
  • External connectivity
  • The fastest, most power-efficient embedded AI computing device: NVIDIA JETSON TX2
  • Hardware-accelerated Computer Vision and AI Libraries
  • API for communication and hardware control
  • Ready-to-use, pre-trained neural networks or easily plug your own in
  • Capability to import neural networks in most common formats (caffe, darknet and soon tensorrt)
  • A complete, plug-in based, application framework
  • Development toolchain
  • Updates
  • Support tickets

Commercial or deployment licenses (except for the Education package).

What is the difference between developer license and deployment license? Which license should I choose?

You will need one developer license for each Pensar that you use to develop your own application. Several developers can share a single Pensar under a single development license. If you need more than one Pensar to develop your own application, you will need a development license per Pensar.
Once your application for Pensar is ready and you need to deploy it on field, you will need to purchase a deployment license per Pensar.
I.e. I am a software consultant company and I need to develop a computer vision, ai-based application running on an embedded device for customer X. I have two developers and I want each of them to have a Pensar and an SDK to be able to develop the application I need. I purchase two Pensars with two development licenses for the SDK.
Other two developers onboard but they use the existing two Pensar using them in turns. I do not need to purchase any additional license.
Once my software is ready, customer X wants to deploy a Pensar per each of the 10 sites he owns. I then need to purchase ten Pensar cameras plus ten deployment licenses.

What is the difference between standard and premium packages?

The main difference between those packages are the support tickets and response time.
Standard package: Limited to 8 support tickets, response time: 48h.
Premium package: Unlimited number of support tickets, response time: 24h.

What is a support ticket?

The support portal runs tickets managing system where the user can ask question that normally fall into the following categories:
  1. The user is correctly using the SDK but some part of it crashes, does not work correctly or does not work as described. This is a bug ticket and your number of support ticket will not be impacted by these requests. Problem will be solved through a new SDK software release.
  2. The user is not able to use a feature or thinks that there is a problem in the SDK but there is issues with the SDK. This is a support ticket. These tickets are limited to the plan included in the purchased license. If you need more, you have to upgrade to the upper package.
  3. The user requests a feature that is not included in the SDK (i.e. support different hardware or cameras or include new neural networks). The feature can or cannot be already planned. This is a feature request ticket and must be evaluated on a one-by-one basis. Under no circumstances, the support tickets will include helping the user develop its own application, reviewing its source code or giving implementation suggestions. If needed, this requests must be handled as a service and be charged on a on-demand basis.

What happens when I used all my support tickets in the Standard package?

You will have to upgrade to the Premium package in order to be able to submit new support tickets.

Do we access, collect, store or use your data?

Absolutely not. We do not have access to the images, videos or metadata that you store on Pensar or you collect with it.

Do I need specific skills to develop AI applications with Pensar SDK?

You will need to be familiar with Python 3 and C++ programming languages. The SDK makes it easier to develop application also for beginner programmers. If you want to add a Graphical User Interface to your plug-in, you will need to get familiar with Kivy .kv language (more information here: https://kivy.org/doc/stable/guide/lang.html).
Please notice that Pensar SDK do not offer any tools to develop your own neural network or train an existing one with your data. In order to do that, you will have to rely on third-party tools and framework. However, Pensar SDK make it easy to import such network into Pensar and to create a computer-vision application around them.

What equipment do I need to develop AI applications?

The minimal equipment is just a Pensar, an HDMI display, usb or bluetooth keyboard and mouse and a wifi internet connection (only needed to download updates).
Pensar SDK includes also tool to develop your application on your own laptop or workstation, and then deploy it on Pensar. This is recommended for a more comfortable developer environment. Notice that a compatible NVIDIA graphic card is required (CUDA SDK 10.0 – 10.1 support for compute capability 3.0 – 7.5 (Kepler, Maxwell, Pascal, Volta, Turing)).
Pensar cannot be used to perform neural network training and Pensar SDK does not include any tool for this.

How much time is it going to take to develop AI applications?

Pensar SDK provides you with a full application framework ready to use and based on a plug-in structure. If your application can be realized with a combination/customization of the provided plug-ins and/or neural networks, you can be ready in days.
If you need to develop your own plug-in, additional development time will be needed depending on the complexity of your project and the size of your team. Pensar SDK includes all the needed tools and a simple workflow to make this process as rapid as possible.
Please notice that Pensar SDK has no tool to help you develop or train your own neural network, so you will have to take into account additional time for that in case you need it.

Which frameworks are included?

AI hardware accelerated libraries provided with Pensar middleware include:
  • NVIDIA cuDnn 7
  • NVIDIA TensorRT 4
  • NVIDIA Caffe
  • OpenCV 3.4.5 (CUDA and ARM Neon acceleration supported)
  • DLIB (CUDA and ARM Neon acceleration supported)
  • GStreamer with NVIDIA proprietary CUDA accelerated plugins
  • Darknet (CUDA accelerated)
TensorRT, Caffe, OpenCV, DLIB and Darknet make possible to load and run the most common AI neural network model formats that include:
  • TensorFlow
  • Caffe
  • ONNX
  • Darknet networks (like for instance YOLO neural networks)


Other libraries provided with Pensar middleware include:
  • Keras
  • Theano
  • TensorFlow
  • SciKit
  • SciPy
  • Numpy
  • Matplotlib
  • ZMQ
  • Mavlink
  • Kivy (used for Graphical User Interfaces)
  • Docker-ce
  • Visual Studio Code

What are the neural networks included?

Latest versions of Yolo (v2, v3 and v2-tiny), ResNet, MobileNetSSD, GoogleNet, Dlib FHogFaceDetector, Dlib Frontal Face Detector, Dlib ANetFaceEncoder, adapted for fast inference on GPU.

What is the warranty for the Pensar camera?

There is 1 year warranty for the Pensar camera.