Method for determining the confidence of a disparity map

The invention relates to a method and a sensor system, designed in particular for determining the confidence of disparity maps inferred by a stereo algorithm or a network through a neural network capable of self-adapting.

Patent title Method for determining the confidence of a disparity map through a self-adaptive learning of a neural network, and sensor system thereof
Thematic area Industry, Digital and Security
Ownership ALMA MATER STUDIORUM - UNIVERSITA' DI BOLOGNA
Inventors Fabio Tosi, Matteo Poggi, Filippo Aleotti, Stefano Mattoccia
Protection Italy
Licensing status Available for development agreements, option, license and other exploitation agreements
Keywords Depth estimation, Uncertainty estimation, Confidence estimation, Machine learning, Deep learning, Self-adapting method
Filed on 02 July 2020

There are in the market several systems for acquiring images in 3D, in order to determine the depth of an image. Currently, stereo is one of the most popular strategies to accurately perceive the 3D structure of the scene, through synchronized cameras and several algorithms. In many practical applications, alongside with disparity inference, confidence estimation is oftenperformed as well. Purposely, a wide range of methods based either on hand-crafted measures or learning-based strategies have been proposed. 

The invention relates to a method and a sensor system, designed in particular for determining the confidence of disparity maps inferred by a stereo algorithm or a network through a neural network capable of self-adapting, but which can be used for any type of image acquisition system, in which it is necessary to estimate the confidence, thus determining the level of certainty or uncertainty of each pixel of said image. 

APPLICATIONS:

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

 

ADVANTAGES:

  • Method for self-adapting a confidence measure unconstrained to the stereo system deployed.
  • High reliability;
  • Easy to implement;
  • Competitive in terms of costs.
Page published on: 30 November 2020