Variational Autoencoder for 3D Voxel Compression. Liu, J., Mills, S., & McCane, B. In 2020 35th International Conference on Image and Vision Computing New Zealand (IVCNZ), pages 1-6, 11, 2020.
![Variational Autoencoder for 3D Voxel Compression [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper doi abstract bibtex
Paper doi abstract bibtex 3D scene sensing and understanding is a fundamental task in the field of computer vision and robotics. One widely used representation for 3D data is a voxel grid. However, explicit representation of 3D voxels always requires large storage space, which is not suitable for light-weight applications and scenarios such as robotic navigation and exploration. In this paper we propose a method to compress 3D voxel grids using an octree representation and Variational Autoencoders (VAEs). We first capture a 3D voxel grid -in our application with collaborating Realsense D435 and T265 cameras. The voxel grid is decomposed into three types of octants which are then compressed by the encoder and reproduced by feeding the latent code into the decoder. We demonstrate the efficiency of our method by two applications: scene reconstruction and path planning.
@inproceedings{
title = {Variational Autoencoder for 3D Voxel Compression},
type = {inproceedings},
year = {2020},
keywords = {Computational modeling,Data models,Image reconstruction,Octrees,Solid modeling,Task analysis,Three-dimensional displays},
pages = {1-6},
month = {11},
id = {9de0e1bc-0393-301b-9e34-a37a2fcfd82b},
created = {2022-03-28T09:45:02.106Z},
file_attached = {true},
profile_id = {235249c2-3ed4-314a-b309-b1ea0330f5d9},
group_id = {1ff583c0-be37-34fa-9c04-73c69437d354},
last_modified = {2022-03-29T08:03:26.589Z},
read = {false},
starred = {false},
authored = {false},
confirmed = {true},
hidden = {false},
citation_key = {liuVariationalAutoencoder3D2020a},
source_type = {inproceedings},
notes = {ISSN: 2151-2205},
private_publication = {false},
abstract = {3D scene sensing and understanding is a fundamental task in the field of computer vision and robotics. One widely used representation for 3D data is a voxel grid. However, explicit representation of 3D voxels always requires large storage space, which is not suitable for light-weight applications and scenarios such as robotic navigation and exploration. In this paper we propose a method to compress 3D voxel grids using an octree representation and Variational Autoencoders (VAEs). We first capture a 3D voxel grid -in our application with collaborating Realsense D435 and T265 cameras. The voxel grid is decomposed into three types of octants which are then compressed by the encoder and reproduced by feeding the latent code into the decoder. We demonstrate the efficiency of our method by two applications: scene reconstruction and path planning.},
bibtype = {inproceedings},
author = {Liu, Juncheng and Mills, Steven and McCane, Brendan},
doi = {10.1109/IVCNZ51579.2020.9290656},
booktitle = {2020 35th International Conference on Image and Vision Computing New Zealand (IVCNZ)}
}Downloads: 0
{"_id":"ZgKghr6aQXx5AQkFs","bibbaseid":"liu-mills-mccane-variationalautoencoderfor3dvoxelcompression-2020","author_short":["Liu, J.","Mills, S.","McCane, B."],"bibdata":{"title":"Variational Autoencoder for 3D Voxel Compression","type":"inproceedings","year":"2020","keywords":"Computational modeling,Data models,Image reconstruction,Octrees,Solid modeling,Task analysis,Three-dimensional displays","pages":"1-6","month":"11","id":"9de0e1bc-0393-301b-9e34-a37a2fcfd82b","created":"2022-03-28T09:45:02.106Z","file_attached":"true","profile_id":"235249c2-3ed4-314a-b309-b1ea0330f5d9","group_id":"1ff583c0-be37-34fa-9c04-73c69437d354","last_modified":"2022-03-29T08:03:26.589Z","read":false,"starred":false,"authored":false,"confirmed":"true","hidden":false,"citation_key":"liuVariationalAutoencoder3D2020a","source_type":"inproceedings","notes":"ISSN: 2151-2205","private_publication":false,"abstract":"3D scene sensing and understanding is a fundamental task in the field of computer vision and robotics. One widely used representation for 3D data is a voxel grid. However, explicit representation of 3D voxels always requires large storage space, which is not suitable for light-weight applications and scenarios such as robotic navigation and exploration. In this paper we propose a method to compress 3D voxel grids using an octree representation and Variational Autoencoders (VAEs). We first capture a 3D voxel grid -in our application with collaborating Realsense D435 and T265 cameras. The voxel grid is decomposed into three types of octants which are then compressed by the encoder and reproduced by feeding the latent code into the decoder. We demonstrate the efficiency of our method by two applications: scene reconstruction and path planning.","bibtype":"inproceedings","author":"Liu, Juncheng and Mills, Steven and McCane, Brendan","doi":"10.1109/IVCNZ51579.2020.9290656","booktitle":"2020 35th International Conference on Image and Vision Computing New Zealand (IVCNZ)","bibtex":"@inproceedings{\n title = {Variational Autoencoder for 3D Voxel Compression},\n type = {inproceedings},\n year = {2020},\n keywords = {Computational modeling,Data models,Image reconstruction,Octrees,Solid modeling,Task analysis,Three-dimensional displays},\n pages = {1-6},\n month = {11},\n id = {9de0e1bc-0393-301b-9e34-a37a2fcfd82b},\n created = {2022-03-28T09:45:02.106Z},\n file_attached = {true},\n profile_id = {235249c2-3ed4-314a-b309-b1ea0330f5d9},\n group_id = {1ff583c0-be37-34fa-9c04-73c69437d354},\n last_modified = {2022-03-29T08:03:26.589Z},\n read = {false},\n starred = {false},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n citation_key = {liuVariationalAutoencoder3D2020a},\n source_type = {inproceedings},\n notes = {ISSN: 2151-2205},\n private_publication = {false},\n abstract = {3D scene sensing and understanding is a fundamental task in the field of computer vision and robotics. One widely used representation for 3D data is a voxel grid. However, explicit representation of 3D voxels always requires large storage space, which is not suitable for light-weight applications and scenarios such as robotic navigation and exploration. In this paper we propose a method to compress 3D voxel grids using an octree representation and Variational Autoencoders (VAEs). We first capture a 3D voxel grid -in our application with collaborating Realsense D435 and T265 cameras. The voxel grid is decomposed into three types of octants which are then compressed by the encoder and reproduced by feeding the latent code into the decoder. We demonstrate the efficiency of our method by two applications: scene reconstruction and path planning.},\n bibtype = {inproceedings},\n author = {Liu, Juncheng and Mills, Steven and McCane, Brendan},\n doi = {10.1109/IVCNZ51579.2020.9290656},\n booktitle = {2020 35th International Conference on Image and Vision Computing New Zealand (IVCNZ)}\n}","author_short":["Liu, J.","Mills, S.","McCane, B."],"urls":{"Paper":"https://bibbase.org/service/mendeley/bfbbf840-4c42-3914-a463-19024f50b30c/file/653e167b-8c61-4e1f-5414-cae1c779d34e/Liu_et_al___2020___Variational_Autoencoder_for_3D_Voxel_Compression.pdf.pdf"},"biburl":"https://bibbase.org/service/mendeley/bfbbf840-4c42-3914-a463-19024f50b30c","bibbaseid":"liu-mills-mccane-variationalautoencoderfor3dvoxelcompression-2020","role":"author","keyword":["Computational modeling","Data models","Image reconstruction","Octrees","Solid modeling","Task analysis","Three-dimensional displays"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"inproceedings","biburl":"https://bibbase.org/service/mendeley/bfbbf840-4c42-3914-a463-19024f50b30c","dataSources":["2252seNhipfTmjEBQ"],"keywords":["computational modeling","data models","image reconstruction","octrees","solid modeling","task analysis","three-dimensional displays"],"search_terms":["variational","autoencoder","voxel","compression","liu","mills","mccane"],"title":"Variational Autoencoder for 3D Voxel Compression","year":2020}