TY - JOUR
T1 - Adversarial Confidence Estimation Networks for Robust Stereo Matching
AU - Kim, Sunok
AU - Min, Dongbo
AU - Kim, Seungryong
AU - Sohn, Kwanghoon
N1 - Publisher Copyright:
© 2000-2011 IEEE.
PY - 2021/11/1
Y1 - 2021/11/1
N2 - Stereo matching aiming to perceive the 3-D geometry of a scene facilitates numerous computer vision tasks used in advanced driver assistance systems (ADAS). Although numerous methods have been proposed for this task by leveraging deep convolutional neural networks (CNNs), stereo matching still remains an unsolved problem due to its inherent matching ambiguities. To overcome these limitations, we present a method for jointly estimating disparity and confidence from stereo image pairs through deep networks. We accomplish this through a minmax optimization to learn the generative cost aggregation networks and discriminative confidence estimation networks in an adversarial manner. Concretely, the generative cost aggregation networks are trained to accurately generate disparities at both confident and unconfident pixels from an input matching cost that are indistinguishable by the discriminative confidence estimation networks, while the discriminative confidence estimation networks are trained to distinguish the confident and unconfident disparities. In addition, to fully exploit complementary information of matching cost, disparity, and color image in confidence estimation, we present a dynamic fusion module. Experimental results show that this model outperforms the state-of-The-Art methods on various benchmarks including real driving scenes.
AB - Stereo matching aiming to perceive the 3-D geometry of a scene facilitates numerous computer vision tasks used in advanced driver assistance systems (ADAS). Although numerous methods have been proposed for this task by leveraging deep convolutional neural networks (CNNs), stereo matching still remains an unsolved problem due to its inherent matching ambiguities. To overcome these limitations, we present a method for jointly estimating disparity and confidence from stereo image pairs through deep networks. We accomplish this through a minmax optimization to learn the generative cost aggregation networks and discriminative confidence estimation networks in an adversarial manner. Concretely, the generative cost aggregation networks are trained to accurately generate disparities at both confident and unconfident pixels from an input matching cost that are indistinguishable by the discriminative confidence estimation networks, while the discriminative confidence estimation networks are trained to distinguish the confident and unconfident disparities. In addition, to fully exploit complementary information of matching cost, disparity, and color image in confidence estimation, we present a dynamic fusion module. Experimental results show that this model outperforms the state-of-The-Art methods on various benchmarks including real driving scenes.
KW - Stereo confidence
KW - confidence estimation
KW - dynamic feature fusion
KW - generative adversarial network
UR - http://www.scopus.com/inward/record.url?scp=85118828736&partnerID=8YFLogxK
U2 - 10.1109/TITS.2020.2995996
DO - 10.1109/TITS.2020.2995996
M3 - Article
AN - SCOPUS:85118828736
SN - 1524-9050
VL - 22
SP - 6875
EP - 6889
JO - IEEE Transactions on Intelligent Transportation Systems
JF - IEEE Transactions on Intelligent Transportation Systems
IS - 11
ER -