Microsoft has already launched the new Azure Kinect to replace the well-known Kinect 2 in 2013. Although the first version of kinect was originally released to provide the Xbox console with new ways to play, the real impact that this type of device offered was not in the world of video games. In this way, its true potential was exploited in countless applications in the world of medicine or architecture.

So, with this new version of Kinect, Microsoft wanted to focus on this type of market away from video games. To this end, it has provided its new device and development tools with the most popular functions and improved those already existing in the first versions. The main components of the new Azure Kinect are described below:

  1. 1-MP depth sensor with wide and narrow field-of-view (FOV) options that help you optimize for your application
  2. 7-microphone array for far-field speech and sound capture
  3. 12-MP RGB video camera for an additional color stream that’s aligned to the depth stream
  4. Accelerometer and gyroscope (IMU) for sensor orientation and spatial tracking
  5. External sync pins to easily synchronize sensor streams from multiple Kinect devices

What's inside kinect (Figure by Microsoft)

What’s inside kinect (Figure by Microsoft)

One of the strengths of this device is the facility given to the developers to squeeze the maximum out of the possibilities offered by this new Kinect. For this purpose Microsoft has launched a new developer kit, the Azure Kinect DK. One of the strong points of this new developer kit is that it is focused on integrating very easily with Microsoft’s Azure services, allowing developers to vitaminize their applications. This time the platform has been divided into different SDKs specialized in supporting different Kinect functionalities:

  • Sensor SDK for low-level sensor and device access.
  • Body Tracking SDK for tracking bodies in 3D.
  • Speech Cognitive Services SDK for enabling microphone access and Azure cloud-based speech services.

Azure Kinect DK architecture (Figure by Microsoft)

Hardware requirements

Regarding the hardware requirements to be able to use this developer kit without problems, there are two types, some less demanding to control the basic functionalities and others more demanding to carry out body tracking tasks. However, it is important to point out that this time Microsoft has decided to provide developers with official support for linux distributions such as Ubuntu. The following are the minimum requirements recommended by Microsoft for running the Sensor SDK. Please note, however, that these may vary depending on the configuration used in the device.

  • Seventh Gen Intel® CoreTM i3 Processor (Dual Core 2.4 GHz with HD620 GPU or faster)
  • 4 GB Memory
  • Dedicated USB3 port
  • Graphics driver support for OpenGL 4.4 or DirectX 11.0

To use the body tracking functionality provided in the Body Tracking SDK, the hardware specifications are a little more demanding due to the computational load required for this type of task. The following are Microsft’s recommended requirements:

  • Seventh Gen Intel® CoreTM i5 Processor (Quad Core 2.4 GHz or faster)
  • 4 GB Memory
  • NVIDIA GEFORCE GTX 1070 or better
  • Dedicated USB3 port

For more information, visit the official documentation of Azure Kinect DK.