DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Namil | ko |
dc.contributor.author | Choi, Yukyung | ko |
dc.contributor.author | Hwang, Soonmin | ko |
dc.contributor.author | Kweon, In-So | ko |
dc.date.accessioned | 2018-02-21T05:20:42Z | - |
dc.date.available | 2018-02-21T05:20:42Z | - |
dc.date.created | 2017-11-29 | - |
dc.date.created | 2017-11-29 | - |
dc.date.created | 2017-11-29 | - |
dc.date.issued | 2018-02 | - |
dc.identifier.citation | 32nd AAAI Conference on Artificial Intelligence / 30th Innovative Applications of Artificial Intelligence Conference / 8th AAAI Symposium on Educational Advances in Artificial Intelligence, pp.6983 - 6991 | - |
dc.identifier.uri | http://hdl.handle.net/10203/239993 | - |
dc.description.abstract | To understand the real-world, it is essential to perceive in all-day conditions including cases which are not suitable for RGB sensors, especially at night. Beyond these limitations, the innovation introduced here is a multispectral solution in the form of depth estimation from a thermal sensor without an additional depth sensor. Based on an analysis of multi-spectral properties and the relevance to depth predictions, we propose an efficient and novel multi-task framework called the Multispectral Transfer Network (MTN) to estimate a depth image from a single thermal image. By exploiting geometric priors and chromaticity clues, our model can generate a pixel-wise depth image in an unsupervised manner. Moreover, we propose a new type of multitask module called Interleaver as a means of incorporating the chromaticity and fine details of skip-connections into the depth estimation framework without sharing feature layers. Lastly, we explain a novel technical means of stably training and covering large disparities and extending thermal images to data-driven methods for all-day conditions. In experiments, we demonstrate the better performance and generalization of depth estimation through the proposed multispectral stereo dataset, including various driving conditions. | - |
dc.language | English | - |
dc.publisher | Association for the Advancement of Artificial Intelligence (AAAI) | - |
dc.title | Multispectral Transfer Network: Unsupervised Depth Estimation for All-day Vision | - |
dc.type | Conference | - |
dc.identifier.wosid | 000485488907007 | - |
dc.identifier.scopusid | 2-s2.0-85043493809 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 6983 | - |
dc.citation.endingpage | 6991 | - |
dc.citation.publicationname | 32nd AAAI Conference on Artificial Intelligence / 30th Innovative Applications of Artificial Intelligence Conference / 8th AAAI Symposium on Educational Advances in Artificial Intelligence | - |
dc.identifier.conferencecountry | US | - |
dc.identifier.conferencelocation | Hilton New Orleans Riverside | - |
dc.contributor.localauthor | Kweon, In-So | - |
dc.contributor.nonIdAuthor | Kim, Namil | - |
dc.contributor.nonIdAuthor | Choi, Yukyung | - |
dc.contributor.nonIdAuthor | Hwang, Soonmin | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.