Benchmarking visual SLAM methods in mirror environments
Abstract Visual simultaneous localisation and mapping (vSLAM) finds applications for indoor and outdoor navigation that routinely subjects it to visual complexities, particularly mirror reflections. The effect of mirror presence (time visible and its average size in the frame) was hypothesised to im...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2024-01-01
|
| Series: | Computational Visual Media |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s41095-022-0329-x |
| _version_ | 1797363006875631616 |
|---|---|
| author | Peter Herbert Jing Wu Ze Ji Yu-Kun Lai |
| author_facet | Peter Herbert Jing Wu Ze Ji Yu-Kun Lai |
| author_sort | Peter Herbert |
| collection | DOAJ |
| description | Abstract Visual simultaneous localisation and mapping (vSLAM) finds applications for indoor and outdoor navigation that routinely subjects it to visual complexities, particularly mirror reflections. The effect of mirror presence (time visible and its average size in the frame) was hypothesised to impact localisation and mapping performance, with systems using direct techniques expected to perform worse. Thus, a dataset, MirrEnv, of image sequences recorded in mirror environments, was collected, and used to evaluate the performance of existing representative methods. RGBD ORB-SLAM3 and BundleFusion appear to show moderate degradation of absolute trajectory error with increasing mirror duration, whilst the remaining results did not show significantly degraded localisation performance. The mesh maps generated proved to be very inaccurate, with real and virtual reflections colliding in the reconstructions. A discussion is given of the likely sources of error and robustness in mirror environments, outlining future directions for validating and improving vSLAM performance in the presence of planar mirrors. The MirrEnv dataset is available at https://doi.org/10.17035/d.2023.0292477898 . |
| first_indexed | 2024-03-08T16:15:27Z |
| format | Article |
| id | doaj.art-9a2cf32f448d4e96a28dbe7fcfbc10d6 |
| institution | Directory Open Access Journal |
| issn | 2096-0433 2096-0662 |
| language | English |
| last_indexed | 2024-03-08T16:15:27Z |
| publishDate | 2024-01-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Computational Visual Media |
| spelling | doaj.art-9a2cf32f448d4e96a28dbe7fcfbc10d62024-01-07T12:39:10ZengSpringerOpenComputational Visual Media2096-04332096-06622024-01-0110221524110.1007/s41095-022-0329-xBenchmarking visual SLAM methods in mirror environmentsPeter Herbert0Jing Wu1Ze Ji2Yu-Kun Lai3School of Computer Science and Informatics, Cardiff UniversitySchool of Computer Science and Informatics, Cardiff UniversitySchool of Engineering, Cardiff UniversitySchool of Computer Science and Informatics, Cardiff UniversityAbstract Visual simultaneous localisation and mapping (vSLAM) finds applications for indoor and outdoor navigation that routinely subjects it to visual complexities, particularly mirror reflections. The effect of mirror presence (time visible and its average size in the frame) was hypothesised to impact localisation and mapping performance, with systems using direct techniques expected to perform worse. Thus, a dataset, MirrEnv, of image sequences recorded in mirror environments, was collected, and used to evaluate the performance of existing representative methods. RGBD ORB-SLAM3 and BundleFusion appear to show moderate degradation of absolute trajectory error with increasing mirror duration, whilst the remaining results did not show significantly degraded localisation performance. The mesh maps generated proved to be very inaccurate, with real and virtual reflections colliding in the reconstructions. A discussion is given of the likely sources of error and robustness in mirror environments, outlining future directions for validating and improving vSLAM performance in the presence of planar mirrors. The MirrEnv dataset is available at https://doi.org/10.17035/d.2023.0292477898 .https://doi.org/10.1007/s41095-022-0329-xvisual simultaneous localisation and mapping (vSLAM)mirrorlocalisationmappingreflectiondataset |
| spellingShingle | Peter Herbert Jing Wu Ze Ji Yu-Kun Lai Benchmarking visual SLAM methods in mirror environments Computational Visual Media visual simultaneous localisation and mapping (vSLAM) mirror localisation mapping reflection dataset |
| title | Benchmarking visual SLAM methods in mirror environments |
| title_full | Benchmarking visual SLAM methods in mirror environments |
| title_fullStr | Benchmarking visual SLAM methods in mirror environments |
| title_full_unstemmed | Benchmarking visual SLAM methods in mirror environments |
| title_short | Benchmarking visual SLAM methods in mirror environments |
| title_sort | benchmarking visual slam methods in mirror environments |
| topic | visual simultaneous localisation and mapping (vSLAM) mirror localisation mapping reflection dataset |
| url | https://doi.org/10.1007/s41095-022-0329-x |
| work_keys_str_mv | AT peterherbert benchmarkingvisualslammethodsinmirrorenvironments AT jingwu benchmarkingvisualslammethodsinmirrorenvironments AT zeji benchmarkingvisualslammethodsinmirrorenvironments AT yukunlai benchmarkingvisualslammethodsinmirrorenvironments |