December 17, 2021

StradVision AR 400x275StradVision, which develops AI-based vision processing technology for autonomous vehicles and advanced driving assistance systems (ADAS), has announced a partnership with LG Electronics to develop an advanced cockpit instrument platform, based on augmented reality.

The companies said automakers are paying more attention to the advantages of AR solutions, particularly augmented reality head-up displays (AR-HUD). The technology projects important road information such as navigation and various ADAS warnings into the line of sight of drivers, allowing them to keep their eyes on the road while also perceiving the surroundings accurately through real-time interaction with objects in their path.

The StradVision technology for this application is delivered by its SVNet camera perception software, and will focus on improving visuals for head-up displays or center screens of automobiles. The software provides the most enhanced vision perception features, including:

  • Depth-map solution: Implements the Pseudo Lidar technology, which replaces high-cost and high-performance solution, this offers the high precision of distance measurement to an object with only a mono-channel camera.
  • Semantic Segmentation: A technology that classifies objects by analyzing the images acquired through the vehicle’s camera on a pixel-by-pixel basis, through deep learning technology.
  • Lane detection: Recognizing the lines on the road for essential ADAS features, such as lane departure warning, lane keeping assistance systems, and lane centering assist.

SVNet is lightweight software that lets vehicles detect and identify objects accurately, such as other vehicles, lanes, pedestrians, animals, free space, traffic signs, and lights, even in harsh weather conditions or bad lighting. The software uses deep learning-based perception algorithms, which is more compact and requires less memory and electricity than competitors, StradVision said. The SVNet software supports more than 14 hardware platforms, and can also be customized for other hardware systems with its Deep Neural Network-enabled technology. 

For more details on StradVision and its offerings, visit its website here.