A Data-Driven Regularization Model for Stereo and Flow
Data-driven techniques can reliably build semantic correspondence among images. In this paper, we present a new regularization model for stereo or flow through transferring the shape information of the disparity or flow from semantically matched patches in the training database. Compared to previous...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2015
|
| Online Access: | http://hdl.handle.net/1721.1/100257 https://orcid.org/0000-0002-2329-5484 https://orcid.org/0000-0002-2231-7995 |
| _version_ | 1826190480202792960 |
|---|---|
| author | Freeman, William T. Wei, Donglai Liu, Ce |
| author2 | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science |
| author_facet | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Freeman, William T. Wei, Donglai Liu, Ce |
| author_sort | Freeman, William T. |
| collection | MIT |
| description | Data-driven techniques can reliably build semantic correspondence among images. In this paper, we present a new regularization model for stereo or flow through transferring the shape information of the disparity or flow from semantically matched patches in the training database. Compared to previous regularization models based on image appearance alone, we can better resolve local ambiguity of the disparity or flow by considering the semantic information without explicit object modeling. We incorporate this data-driven regularization model into a standard Markov Random Field (MRF) model, inferred with a gradient descent algorithm and learned with a discriminative learning approach. Compared to prior state-of-the-art methods, our full model achieves comparable or better results on the KITTI stereo and flow datasets, and improves results on the Sintel Flow dataset under an online estimation setting. |
| first_indexed | 2024-09-23T08:40:35Z |
| format | Article |
| id | mit-1721.1/100257 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T08:40:35Z |
| publishDate | 2015 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/1002572022-09-30T10:28:08Z A Data-Driven Regularization Model for Stereo and Flow Freeman, William T. Wei, Donglai Liu, Ce Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Wei, Donglai Freeman, William T. Data-driven techniques can reliably build semantic correspondence among images. In this paper, we present a new regularization model for stereo or flow through transferring the shape information of the disparity or flow from semantically matched patches in the training database. Compared to previous regularization models based on image appearance alone, we can better resolve local ambiguity of the disparity or flow by considering the semantic information without explicit object modeling. We incorporate this data-driven regularization model into a standard Markov Random Field (MRF) model, inferred with a gradient descent algorithm and learned with a discriminative learning approach. Compared to prior state-of-the-art methods, our full model achieves comparable or better results on the KITTI stereo and flow datasets, and improves results on the Sintel Flow dataset under an online estimation setting. National Science Foundation (U.S.) (CGV 1212849) United States. Office of Naval Research. Multidisciplinary University Research Initiative (Award N00014-09-1-1051) 2015-12-15T02:48:28Z 2015-12-15T02:48:28Z 2014-12 Article http://purl.org/eprint/type/ConferencePaper 978-1-4799-7000-1 http://hdl.handle.net/1721.1/100257 Donglai Wei, Ce Liu, and William T. Freeman. “A Data-Driven Regularization Model for Stereo and Flow.” 2014 2nd International Conference on 3D Vision (December 2014). https://orcid.org/0000-0002-2329-5484 https://orcid.org/0000-0002-2231-7995 en_US http://dx.doi.org/10.1109/3DV.2014.97 Proceedings of the 2014 2nd International Conference on 3D Vision Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Institute of Electrical and Electronics Engineers (IEEE) MIT web domain |
| spellingShingle | Freeman, William T. Wei, Donglai Liu, Ce A Data-Driven Regularization Model for Stereo and Flow |
| title | A Data-Driven Regularization Model for Stereo and Flow |
| title_full | A Data-Driven Regularization Model for Stereo and Flow |
| title_fullStr | A Data-Driven Regularization Model for Stereo and Flow |
| title_full_unstemmed | A Data-Driven Regularization Model for Stereo and Flow |
| title_short | A Data-Driven Regularization Model for Stereo and Flow |
| title_sort | data driven regularization model for stereo and flow |
| url | http://hdl.handle.net/1721.1/100257 https://orcid.org/0000-0002-2329-5484 https://orcid.org/0000-0002-2231-7995 |
| work_keys_str_mv | AT freemanwilliamt adatadrivenregularizationmodelforstereoandflow AT weidonglai adatadrivenregularizationmodelforstereoandflow AT liuce adatadrivenregularizationmodelforstereoandflow AT freemanwilliamt datadrivenregularizationmodelforstereoandflow AT weidonglai datadrivenregularizationmodelforstereoandflow AT liuce datadrivenregularizationmodelforstereoandflow |