Barco Labs (research)

Barco NV
Belgium and Milpitas, CA
April 2017
January 2021

[I know, this section just echos the same stuff as on the résumé. I plan to expand later.] Worked with PhDs, staff and university interns researching disruptive technologies. Barco Labs deliverables are research papers, patents and demos. Any research that might become a viable product in 2 to 5 years is then passed off to one of the product divisions. (Due to the trade secret nature of this research some details cannot be revealed.) Accomplishments:

  • Evaluated and prototyped text segmentation algorithms using DNNs converted from TensorFlow by OpenVINO into Intel Optimized DAG Models for OpenCV 4.3 with Nvidia GPU CUDA acceleration.
  • Researched, created and prototyped “smart” video exposure and white-balance algorithms using colorimetry, OpenCV/Python and ArUco markers for auto-tracking of calibration targets.
  • Prototyped mixed reality mashups using homography and inverse perspective transforms to seamlessly insert overlays into video streams via OpenCV/Python.
  • Implemented video frame motion smoothing using PDI and custom easing functions in C/C++.
  • Implemented a low-latency, multi-stream web video app using OpenTok to convert RTSP videos into WebRTCs packaged by Docker to manage an OpenPose DNN prototype application. (Related patent pending.)
  • Implemented a Progressive Web App (PWA) with native behavior on iOS, Android, Windows and MacOS using Ionic/Angular integrated with EWS (Exchange Web Services) and Leaflet interactive maps.
  • Implemented web UIs showing DNN anomaly detections using Angular-Meteor and WebRTC packaged by Docker with post-playback and event time lines.
  • Researched and prototyped 3-D situational awareness mashups for web delivery using WebAssembly (Emscripten), WebRTC, OpenGL shader programming and the Unreal game engine in C/C++.
  • Created labeled image training sets using the Unity game engine and Python for DNN training related to facial features. (DNN patent pending.)