Method for determining the depth of images

The invention refers to a method for 3D reconstruction using a stereo camera and a neural network based on automatic learning which self- adapts to the different environmental conditions during operation.

Title of the patent Method for determining the depth of images
Thematic area Electrical, electronic and information engineering
Ownership Alma Mater Auditorum - University of Bologna
Inventors Alessio Tonioni, Matteo Poggi, Fabio Tosi, Stefano Mattoccia, Luigi Di Stefano
Protection Italy (with possibility to extent internationally)
Licensing status Available for licensing agreements
Keywords 3D reconstruction, Stereo matching, Deep learning, Self-adaptive, Depth from images
Filed on Dec 02, 2019

The information on the depth of the observed points is crucial and some times essential in applications such as, for example, autonomous or assisted driving, 3D reconstruction starting from 2D images or Augmented Reality.

The patented method reduces the need  for data necessary for the network training before it becomes effective. Compared to deep learning techniques mainly based on Convolutional Neural Network implemented on bulky and/or high-energy consuming devices, they are less expensive in terms of computation. A low-energy-consumption stereo smart camera could be produced on the basis of the invention. It would be able to self-adapt to the examined environment in real time while processing stereo edge maps. 

APPLICATIONS:

  • Autonomous driving systems;
  • Computer Vision;
  • Robotics;
  • Augmented Reality,
  • Ricostruzione 3D.

ADVANTAGES:

  • Self-training system;
  • Accuracy maintained unchanged;
  • Image processing both on-board and on other external hardware devices.