Augmented Reality (AR), Virtual Reality (VR), Mixed Reality (MR), and Extended Reality (XR) technologies have continued to gain popularity in such industries and applications as entertainment, healthcare, automotive, education, etc. However, the productization of AR/VR equipment faces many challenges, as image quality is paramount for acceptable AR and VR experiences.

VR gaming headsets require high-quality images to provide realistic and clear virtual scenes. To improve the user immersion experience, eye tracking using cameras is commonly used to tune the images according to the eye positions. The difficulty comes from requiring high enough contrast for accurate eye detection and high enough camera FPS to track eye movements fast enough.

AR gaming glasses generally integrate Simultaneous Localization and Mapping (SLAM) technology, connecting users to the real world by mapping their surroundings in real time. For this purpose, DXOMARK Analyzer solution offers a precise initial calibration of the camera’s intrinsic and extrinsic parameters.

In healthcare, AR/VR headsets with see-through technology aid in surgery by capturing and displaying the real-world environment using a stereo camera. The reproduction of the real world should be as authentic as possible to achieve high accuracy during the operation. Such cameras generally use short exposure times and small apertures to achieve a large depth of field, resulting in a low number of captured photons. Achieving optimal image quality involves judicious Image Signal Processing (ISP) tuning to find the best tradeoff between noise reduction and texture preservation. Moreover, reducing latency is crucial for matching VR scenery to head movements, which is why DXOMARK Analyzer provides a series of measurements that assesses the performance of cameras in terms of noise reduction, texture preservation, and timing. Analyzer also assesses such key image quality attributes as contrast, color fidelity, HDR capabilities and distortion to enhance high precision during surgeries.

DXOMARK Analyzer solution includes the Multispectral Lighting System (MLS) to simulate diverse lighting conditions by reproducing corresponding light spectra and intensities. This allows for assessing the image quality of AR/VR devices across various use cases.

​Use Cases:
  • SLAM (simultaneous localization and mapping) using multiple cameras. It requires extrinsic and intrinsic calibration of all the cameras.
  • Vision through: a stereocam is used to capture the real world and display it in the headset. The cameras use a short exposure time and a small aperture (to get a high Depth of Field), so the number of captured photons is low. The quality of the images projected needs a lot of ISP tuning to handle the better compromise between noise and texture preservation.
  • Eye tracking with camera: to improve the experience, the headset is using the eye position to tune the images.

 

 

Measurement and KPIs:
  • Image quality: MTF, texture preservation, color fidelity, visual noise, HDR, Distortion.
  • DXOMARK RAW: a good HW is necessary to get good RGB images
  • Calibration 3D
  • Timing and flickering mitigation
  • Flare

 

Starter Kit: