DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Beomyoung | ko |
dc.contributor.author | Han, Sangeun | ko |
dc.contributor.author | Yi, Eojindl | ko |
dc.contributor.author | Kim, Junmo | ko |
dc.date.accessioned | 2021-11-05T06:41:25Z | - |
dc.date.available | 2021-11-05T06:41:25Z | - |
dc.date.created | 2021-10-26 | - |
dc.date.issued | 2021-07 | - |
dc.identifier.citation | 14th International Conference on Multi-disciplinary Trends in Artificial Intelligence, MIWAI 2021, pp.1 - 13 | - |
dc.identifier.issn | 0302-9743 | - |
dc.identifier.uri | http://hdl.handle.net/10203/288882 | - |
dc.description.abstract | Progress in LiDAR sensors have opened up the potential for novel applications using point clouds. However, LiDAR sensors are inherently sensitive, and also lack the ability to colorize point clouds, thus impeding further development of the applications mentioned above. Our paper presents a new end-to-end network that upsamples and colorizes a given input point cloud. Thus the network is able to manage the sparseness and noisiness resulting from the sensitivity of the sensor, and also enrich point cloud data by giving them the original color in the real world. To the best of our knowledge, this is the first work that uses a voxelized generative model to colorize point clouds, and also the first to perform both upsampling and colorization tasks in a single network. Experimental results show that our model is able to correctly colorize and upsample a given input point cloud. From this, we conclude that our model understands the shape and color of various objects. © 2021, Springer Nature Switzerland AG. | - |
dc.language | English | - |
dc.publisher | Springer Science and Business Media Deutschland GmbH | - |
dc.title | 3D Point Cloud Upsampling and Colorization Using GAN | - |
dc.type | Conference | - |
dc.identifier.scopusid | 2-s2.0-85111995504 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 1 | - |
dc.citation.endingpage | 13 | - |
dc.citation.publicationname | 14th International Conference on Multi-disciplinary Trends in Artificial Intelligence, MIWAI 2021 | - |
dc.identifier.conferencecountry | US | - |
dc.identifier.doi | 10.1007/978-3-030-80253-0_1 | - |
dc.contributor.localauthor | Kim, Junmo | - |
dc.contributor.nonIdAuthor | Kim, Beomyoung | - |
dc.contributor.nonIdAuthor | Han, Sangeun | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.