DEEP-LEARNING-BASED STEREO MATCHING METHOD FOR BINOCULAR DYNAMIC VISION SENSOR

A deep-learning-based stereo matching method for a binocular dynamic vision sensor, the method comprising the following steps: 1) generating training point pairs according to depth information in a data set of a binocular event camera; 2) constructing a representation mode suitable for events in an event flow of a dynamic vision sensor; and 3) representing the event training point pairs according to the representation mode, and transmitting same to twin neural networks for training, and performing stereo matching according to a training result. Compared with the prior art, the method has the advantages of high matching precision and a fast matching speed. L'invention concerne un procédé d'appariement stéréoscopique basé sur un apprentissage profond pour un capteur de vision dynamique binoculaire, le procédé comportant les étapes consistant à: 1) générer des paires de points d'apprentissage d'après des informations de profondeur dans un jeu de données d'une caméra binoculaire d'événements; 2) construire un mod