Practical deep stereo (PDS): Toward applications-friendly deep stereo matching

Stepan Tulyakov, Anton Ivanov, Francois Fleuret

Research output: Contribution to journalConference articlepeer-review

59 Citations (Scopus)


End-to-end deep-learning networks recently demonstrated extremely good performance for stereo matching. However, existing networks are difficult to use for practical applications since (1) they are memory-hungry and unable to process even modest-size images, (2) they have to be trained for a given disparity range. The Practical Deep Stereo (PDS) network that we propose addresses both issues: First, its architecture relies on novel bottleneck modules that drastically reduce the memory footprint in inference, and additional design choices allow to handle greater image size during training. This results in a model that leverages large image context to resolve matching ambiguities. Second, a novel sub-pixel cross-entropy loss combined with a MAP estimator make this network less sensitive to ambiguous matches, and applicable to any disparity range without re-training. We compare PDS to state-of-the-art methods published over the recent months, and demonstrate its superior performance on FlyingThings3D and KITTI sets.

Original languageEnglish
Pages (from-to)5871-5881
Number of pages11
JournalAdvances in Neural Information Processing Systems
Publication statusPublished - 2018
Externally publishedYes
Event32nd Conference on Neural Information Processing Systems, NeurIPS 2018 - Montreal, Canada
Duration: 2 Dec 20188 Dec 2018


Dive into the research topics of 'Practical deep stereo (PDS): Toward applications-friendly deep stereo matching'. Together they form a unique fingerprint.

Cite this