Publication
Deep Learning-Based Point Cloud Geometry Coding: RD Control Through Implicit and Explicit Quantization
| datacite.subject.sdg | 03:Saúde de Qualidade | |
| datacite.subject.sdg | 10:Reduzir as Desigualdades | |
| datacite.subject.sdg | 11:Cidades e Comunidades Sustentáveis | |
| dc.contributor.author | Guarda, André F. R. | |
| dc.contributor.author | Rodrigues, Nuno M. M. | |
| dc.contributor.author | Pereira, Fernando | |
| dc.date.accessioned | 2025-09-09T13:43:06Z | |
| dc.date.available | 2025-09-09T13:43:06Z | |
| dc.date.issued | 2020-07 | |
| dc.description | Article number - 9106022; Conference city - London; Conference date - 6 July 2020 - 10 July 2020; Conference code - 162345 | |
| dc.description | EISBN - 978-1-7281-1485-9 | |
| dc.description.abstract | Deep learning is becoming more and more relevant for multiple multimedia processing tasks, and lately it has raised much interest in the coding arena notably for images and point clouds. While offering near state-of-the-art compression performance, current deep learning-based point cloud coding solutions have a shortcoming since they require training and storing multiple models in order to obtain different rate-distortion trade-offs. This paper proposes a solution that effectively reduces the number of deep learning models that need to be trained and stored by applying explicit quantization to the latent representation, which can be controlled at coding time, to generate varying rate-distortion tradeoffs. The proposed implicit-explicit quantization combination achieves a compression performance that is equivalent or better than the alternative, while significantly reducing the model storage memory requirements. | eng |
| dc.description.sponsorship | This work was funded by Fundação para a Ciência e Tecnologia (FCT), Portugal, Ph.D. Grant SFRH/BD/ 118218/2016, by FCT/MEC through national funds and when applicable co-funded by EU funds under the project UIDB/EEA/50008/2020. | |
| dc.identifier.citation | A. F. R. Guarda, N. M. M. Rodrigues and F. Pereira, "Deep Learning-Based Point Cloud Geometry Coding: RD Control Through Implicit and Explicit Quantization," 2020 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), London, UK, 2020, pp. 1-6, doi: https://doi.org/10.1109/ICMEW46912.2020.9106022. | |
| dc.identifier.doi | 10.1109/icmew46912.2020.9106022 | |
| dc.identifier.isbn | 978-1-7281-1486-6 | |
| dc.identifier.isbn | 978-1-7281-1485-9 | |
| dc.identifier.uri | http://hdl.handle.net/10400.8/14020 | |
| dc.language.iso | eng | |
| dc.peerreviewed | yes | |
| dc.publisher | IEEE Canada | |
| dc.relation | Efficient lossy and lossless compression of point clouds | |
| dc.relation.hasversion | https://ieeexplore.ieee.org/document/9106022 | |
| dc.relation.ispartof | 2020 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) | |
| dc.rights.uri | N/A | |
| dc.subject | Point cloud coding | |
| dc.subject | deep learning | |
| dc.subject | explicit quantization | |
| dc.subject | multiple models | |
| dc.title | Deep Learning-Based Point Cloud Geometry Coding: RD Control Through Implicit and Explicit Quantization | eng |
| dc.type | conference paper | |
| dspace.entity.type | Publication | |
| oaire.awardTitle | Efficient lossy and lossless compression of point clouds | |
| oaire.awardURI | http://hdl.handle.net/10400.8/12910 | |
| oaire.citation.conferenceDate | 2020-07 | |
| oaire.citation.conferencePlace | London, UK | |
| oaire.citation.title | 2020 IEEE International Conference on Multimedia and Expo Workshops, ICMEW 2020 | |
| oaire.fundingStream | OE | |
| oaire.version | http://purl.org/coar/version/c_970fb48d4fbd8a85 | |
| person.familyName | M. M. Rodrigues | |
| person.givenName | Nuno | |
| person.identifier.orcid | 0000-0001-9536-1017 | |
| person.identifier.scopus-author-id | 7006052345 | |
| relation.isAuthorOfPublication | b4ebe652-7f0e-4e67-adb0-d5ea29fc9e69 | |
| relation.isAuthorOfPublication.latestForDiscovery | b4ebe652-7f0e-4e67-adb0-d5ea29fc9e69 | |
| relation.isProjectOfPublication | d619b8c6-7ef9-4635-98fd-a9a7be25e5f8 | |
| relation.isProjectOfPublication.latestForDiscovery | d619b8c6-7ef9-4635-98fd-a9a7be25e5f8 |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- Deep learning-based point cloud geometry coding RD control through implicit and explicit quantization.pdf
- Size:
- 725.43 KB
- Format:
- Adobe Portable Document Format
- Description:
- Deep learning is becoming more and more relevant for multiple multimedia processing tasks, and lately it has raised much interest in the coding arena notably for images and point clouds. While offering near state-of-the-art compression performance, current deep learning-based point cloud coding solutions have a shortcoming since they require training and storing multiple models in order to obtain different rate-distortion trade-offs. This paper proposes a solution that effectively reduces the number of deep learning models that need to be trained and stored by applying explicit quantization to the latent representation, which can be controlled at coding time, to generate varying rate-distortion tradeoffs. The proposed implicit-explicit quantization combination achieves a compression performance that is equivalent or better than the alternative, while significantly reducing the model storage memory requirements.
License bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- license.txt
- Size:
- 1.32 KB
- Format:
- Item-specific license agreed upon to submission
- Description: