3D Scene Segmentation: A Comprehensive Survey and Open Problems
This paper presents a detailed review of recent advancements in 3D indoor scene segmentation driven by deep learning techniques. It provides an overview of existing segmentation models, examines various data representations, data collection methods, augmentation techniques, and available datasets. A...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11050424/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849416270463631360 |
|---|---|
| author | Slavcho Neshev Krasimir Tonchev Agata Manolova Vladimir Poulkov |
| author_facet | Slavcho Neshev Krasimir Tonchev Agata Manolova Vladimir Poulkov |
| author_sort | Slavcho Neshev |
| collection | DOAJ |
| description | This paper presents a detailed review of recent advancements in 3D indoor scene segmentation driven by deep learning techniques. It provides an overview of existing segmentation models, examines various data representations, data collection methods, augmentation techniques, and available datasets. A comparative analysis of loss functions and overview of evaluation metrics is conducted to highlight their impact on segmentation performance. Unlike previous surveys, this work introduces a new classification of data augmentation techniques and proposes two novel classification approaches for 3D instance and semantic segmentation. Furthermore, it unifies 3D semantic instance segmentation and 3D panoptic segmentation within an existing framework. The paper also identifies key challenges and open research directions, providing insights into future advancements in the field. |
| format | Article |
| id | doaj-art-cae7fec0a5cd4bfe940bc51baa54d6fd |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-cae7fec0a5cd4bfe940bc51baa54d6fd2025-08-20T03:33:14ZengIEEEIEEE Access2169-35362025-01-011311045711049610.1109/ACCESS.2025.3583136110504243D Scene Segmentation: A Comprehensive Survey and Open ProblemsSlavcho Neshev0https://orcid.org/0009-0009-9633-4980Krasimir Tonchev1https://orcid.org/0000-0002-3332-666XAgata Manolova2https://orcid.org/0000-0002-8120-363XVladimir Poulkov3https://orcid.org/0000-0003-3226-5639Faculty of Telecommunications, Technical University of Sofia, Sofia, BulgariaFaculty of Telecommunications, Technical University of Sofia, Sofia, BulgariaFaculty of Telecommunications, Technical University of Sofia, Sofia, BulgariaFaculty of Telecommunications, Technical University of Sofia, Sofia, BulgariaThis paper presents a detailed review of recent advancements in 3D indoor scene segmentation driven by deep learning techniques. It provides an overview of existing segmentation models, examines various data representations, data collection methods, augmentation techniques, and available datasets. A comparative analysis of loss functions and overview of evaluation metrics is conducted to highlight their impact on segmentation performance. Unlike previous surveys, this work introduces a new classification of data augmentation techniques and proposes two novel classification approaches for 3D instance and semantic segmentation. Furthermore, it unifies 3D semantic instance segmentation and 3D panoptic segmentation within an existing framework. The paper also identifies key challenges and open research directions, providing insights into future advancements in the field.https://ieeexplore.ieee.org/document/11050424/3D indoor scene segmentationdeep learningdata acquisitionsegmentation modelsperformance metrics3D models |
| spellingShingle | Slavcho Neshev Krasimir Tonchev Agata Manolova Vladimir Poulkov 3D Scene Segmentation: A Comprehensive Survey and Open Problems IEEE Access 3D indoor scene segmentation deep learning data acquisition segmentation models performance metrics 3D models |
| title | 3D Scene Segmentation: A Comprehensive Survey and Open Problems |
| title_full | 3D Scene Segmentation: A Comprehensive Survey and Open Problems |
| title_fullStr | 3D Scene Segmentation: A Comprehensive Survey and Open Problems |
| title_full_unstemmed | 3D Scene Segmentation: A Comprehensive Survey and Open Problems |
| title_short | 3D Scene Segmentation: A Comprehensive Survey and Open Problems |
| title_sort | 3d scene segmentation a comprehensive survey and open problems |
| topic | 3D indoor scene segmentation deep learning data acquisition segmentation models performance metrics 3D models |
| url | https://ieeexplore.ieee.org/document/11050424/ |
| work_keys_str_mv | AT slavchoneshev 3dscenesegmentationacomprehensivesurveyandopenproblems AT krasimirtonchev 3dscenesegmentationacomprehensivesurveyandopenproblems AT agatamanolova 3dscenesegmentationacomprehensivesurveyandopenproblems AT vladimirpoulkov 3dscenesegmentationacomprehensivesurveyandopenproblems |