What Can We Provide?

By combining our team’s twenty years experience on stereo camera, our preparatory global optimization algorithms for depth calculation, infrared active structure light, and motion sensors, we are able to create a depth sensing unit that gives you

20 years experienceon stereo camera

Preparatory GO algorithm for depth

Preparatory GO algorithm for depth

Infrared structure light

high frame rate real-time (over 60 frames/second)

sub-pixel disparity accuracy

working in both indoor and outdoor, dark or bright

sensitive to rotation

ideal for navigation and obstacle avoidance for any robots, drones, automatous vehicles, and VR/AR applications.

Comprehensive System /Framework Support

What makes MYNT EYE really stand out is our SDK. Our SDK makes MYNT EYE support the broadest operating systems and frameworks for your development.

Windows Installation

Linux Installation

ROS Installation

OpenCV Installation

Mac Installation

Android Installation

NVIDIA Jetson platform

NVIDIA CUDA accleration

Flexible and Complete SDK

With our SDK you will be able to get

  • point cloud

  • depth images

  • disparity images

  • corrected stereo

  • original binocular

In addition you can also access

IMU data

device information

image calibration data

With SDK you also have full control of our MYNT EYE camera: you can set

frame rate

motion sensor data rate

image resolution


enable or disable IR

The best part of our SDK is that we have the broadest support of open source SLAMs. We have built the interface and made our MYNT eye camera work with ORB SLAM, OKVIS, VINS, VIORB, Maplab X……and the lists will keep growing.





We leverage NVIDIA GPU and CUDA to accelerate many of our computations, in particularly, the disparity calculation. You will get the highest frame rate real-time depth and point cloud data on NVIDIA GPUs with CUDA acceleration.

We believe NVIDIA Jetson platform and our MYNTeye are the best couples for robot development, that’s why we offer our best integration with NVIDIA Jetson platform. It is very convenient to develop your projects on NVIDIA Jetson platform with MYNTeye. With our SDK for NVIDIA Jetson platform, you will be able to develop all kinds of applications, not limited to VSLAM, to name a few: 3D face recognition, object detection, 3D metrology, ADAS…

3D face recognition

object detection

3D metrology


Experience Now

Here is our MYNTeye SDK 2.0 GitHub page where you can download our SDK, find our online documentation, and access our online support forum. For hardware, software, libraries, tools and developer documentation of TX1, TX2 and other NVIDIA products, feel free to check out NVIDIA Download Center.


Best for Visual SLAM, Depth and Point Clouds Calculation SDK provided

Stereo Camera / IR Structured-Light / Built-in IMU/ Integrated with Open Source SLAMs / CPU and GPU Depth Calculation SW / SDK for Multi-Platforms


Output High Resolution and High Frame-Rate Depth Output Directly

Output Depth Directly / IMU Integrated / IR Structured-Light / RGB image / Higher Resolution and High Frame Rate

MN 01

The MN-01 is an open source vSLAM development platform. It includes a stereo camera, an NVIDIA Jetson TX2, and a rich expansion interface, all on a mobile base. This product enables you to start developing your robot in a super quick way.

Coming soon......

Jetson TX2
Developer Kit

The Jetson TX2 Developer Kit includes the Jetson TX2 module, and is supported by the MYNT Eye SDK.

Hardware Features:

  • >> NVIDIA Pascal Architecture GPU
  • >> Dual-Core Denver 2 64-bit CPU + Quad-CoreARM A57 Complex
  • >> 8 GB L128 bit DDR4 Memory
  • >> 32 GB eMMC 5.1 Flash Storage
  • >> Connectivity to 802.11ac WLAN and Bluetooth-Enabled Devices
  • >> 10/100/1000BASE-T Ethernet
  • >> Power On/Off
  • >> Reset
  • >> Force Recovery
  • >> Video - 4k x 2k 60Hz Encode [HEVC] 4k x 2k 60Hz Decode [12-Bit Support]
  • >> CSI - Up to 6 Cameras [2 lane] CSI2 D-PHY 1.2 [2.5 Gbps/Lane]
  • >> User-DefinedPower Options
  • >> USB 3.0 Type A
  • >> USB 2.0 Micro AB (supports recovery and host mode)
  • >> USB 2.0 Micro AB (supports recovery and host mode)
  • >> HDMI 2.0 / 2x DSI, 2x DP 1.2 / eDP 1.4
  • >> M.2 Key E
  • >> PCI-E x4
  • >> Gigabit Ethernet
  • >> Full-Size SD
  • >> SATA Data and Power
  • >> GPIOs, I2C, I2S, SPI, CAN*
  • >> TTL UART with Flow Control
  • >> Display Expansion Header*
  • >> Camera Expansion Header*

* I/O expansion headers: refer to product documentation for header specification.

NVIDIA Download Center >