Electronics companies ST Microelectronics and EYS3D are revealing the results of their machine-vision collaboration at this week’s CES in Las Vegas.

STM is a Swiss semiconductor company and EYS3D Microelectronics is a Taiwanese fabless semiconductor design house that focuses on end-to-end hardware and software for computer vision, including vision-processing systems-on-chip (SoCs).

They have been collaborating on high-quality machine vision. At CES, they are using live demonstrations to show how stereo video and a depth camera with active-coded infra-red technology can enhance capabilities such as feature recognition and autonomous guidance at mid-to-long working range.

“ST Microelectronics’ advanced image sensors, using proprietary process technologies, offer class-leading pixel size while offering both high sensitivity and low crosstalk,” said James Wang, chief strategy officer at EYS3D. “Such high-performance image sensors at a competitive price point enable us to achieve extremely compact system size while ensuring outstanding machine-vision performance. The strong connection we have established with STM increases our confidence to develop new products that will lead the machine-vision market.”

The CES demonstrations highlight two jointly developed reference designs, the Ref-B6 and Ref-B3 ASV (active stereo vision) video and depth cameras. Both combine the EYS3D CV processor and eSP876 stereo 3D depth-map chipset with STM’s shutter image sensors that provide enhanced near-infra-red (NIR) sensitivity.

The embedded EYS3D chipset enhances object edge detection, optimises depth de-noising, and outputs HD-quality 3D depth data up to 60fps. STM’s image sensors enable the cameras to output data streams in various combinations of video and depth resolution and frame rate for the best quality depth sensing and point-cloud creation.

In addition, optimised lenses, filters and a VCSEL active-IR projector source optimise the infra-red optical path and increase immunity to ambient light noise. A control algorithm turns the IR projector on and off alternately to permit capturing artefact-free greyscale images. Leveraging this hardware design, the Ref-B6 stereo-video camera achieves a 6cm baseline and 85deg (H) by 70deg (V) depth field of view.

Both EYS3D reference designs include a software development kit (SDK) supporting Windows, Linux and Android OS environments with multiple programming languages and wrapper APIs.

“The collaboration with EYS3D Microelectronics, through their expertise in capture, perception understanding and 3D-fusion, offers STM additional business opportunities, use cases and ecosystems addressing demands for stereo vision in applications such as robots, home-automation, home appliances and many others,” said David Maucotel, business line manager at STM. “While the reference designs showcased at CES are using monochromatic sensors, we can already foresee exciting enhancements and further use-cases using the RGB and RGB-IR versions of our sensors.”

• EYS3D launched at CES a computer vision development platform for autonomous robotic applications such as AIoT, smart city, indoor cleaning robots, and outdoor agricultural robots for industrial and retail sectors.

The Xink platform is both a platform-as-a-service (PaaS) as well as a hardware and software development kit for the design of safe, vision-capable mobile AI products equipped with field analysis, object recognition, obstacle detection, object tracking and following, and route planning functions.

Xink provides the necessary elements for product development, including compute power, AI accelerator, IO controls and Flexibus communication peripherals, smart power management and machine vision subsystems. The modular platform takes care of low-level programming, freeing developers to use cut-and-paste coding for application-specific design while reducing design cycles for quicker commercialisation.

“New technologies have enabled surging market growth for artificial intelligence, and semiconductors play the core role,” said Nicky Lu, CEO of EYS3D parent company Etron. “In addition to establishing an AI ecosystem and developing key technologies, we are proud of launching Xink.”