Skip to Main content Skip to Navigation
New interface
Conference papers

Fast Stereo Disparity Maps Refinement By Fusion of Data-Based And Model-Based Estimations

Maxime Ferrera 1, 2 Alexandre Boulch 1 Julien Moras 1 
2 ICAR - Image & Interaction
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : The estimation of disparity maps from stereo pairs has many applications in robotics and autonomous driving. Stereo matching has first been solved using model-based approaches, with real-time considerations for some, but to-day's most recent works rely on deep convolutional neural networks and mainly focus on accuracy at the expense of computing time. In this paper, we present a new method for disparity maps estimation getting the best of both worlds: the accuracy of data-based methods and the speed of fast model-based ones. The proposed approach fuses prior disparity maps to estimate a refined version. The core of this fusion pipeline is a convolutional neural network that leverages dilated convolutions for fast context aggregation without spatial resolution loss. The resulting architecture is both very effective for the task of refining and fusing prior disparity maps and very light, allowing our fusion pipeline to produce disparity maps at rates up to 125 Hz. We obtain state-of-the-art results in terms of speed and accuracy on the KITTI benchmarks. Code and pre-trained models are available on our github: ferreram/FD-Fusion.
Complete list of metadata

Cited literature [27 references]  Display  Hide  Download
Contributor : Maxime Ferrera Connect in order to contact the contributor
Submitted on : Tuesday, October 22, 2019 - 4:06:05 PM
Last modification on : Monday, October 3, 2022 - 1:41:54 PM
Long-term archiving on: : Thursday, January 23, 2020 - 8:55:29 PM


Files produced by the author(s)



Maxime Ferrera, Alexandre Boulch, Julien Moras. Fast Stereo Disparity Maps Refinement By Fusion of Data-Based And Model-Based Estimations. 3DV 2019 - International Conference on 3D Vision, Sep 2019, Québec, Canada. pp.9-17, ⟨10.1109/3DV.2019.00011⟩. ⟨hal-02326896⟩



Record views


Files downloads