Leveraging a weakly adversarial paradigm for joint learning of disparity and confidence estimation.

Matteo Poggi, Fabio Tosi, Filippo Aleotti, Stefano Mattoccia

Browse the full ICPR paper archive.