Self-Supervised Sidewalk Perception Using Fast Video Semantic Segmentation for Robotic Wheelchairs in Smart Mobility
The real-time segmentation of sidewalk environments is critical to achieving autonomous navigation for robotic wheelchairs in urban territories. A robust and real-time video semantic segmentation offers an apt solution for advanced visual perception in such complex domains. The key to this propositi...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-07-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/14/5241 |
| _version_ | 1827618617109250048 |
|---|---|
| author | Vishnu Pradeep Redouane Khemmar Louis Lecrosnier Yann Duchemin Romain Rossi Benoit Decoux |
| author_facet | Vishnu Pradeep Redouane Khemmar Louis Lecrosnier Yann Duchemin Romain Rossi Benoit Decoux |
| author_sort | Vishnu Pradeep |
| collection | DOAJ |
| description | The real-time segmentation of sidewalk environments is critical to achieving autonomous navigation for robotic wheelchairs in urban territories. A robust and real-time video semantic segmentation offers an apt solution for advanced visual perception in such complex domains. The key to this proposition is to have a method with lightweight flow estimations and reliable feature extractions. We address this by selecting an approach based on recent trends in video segmentation. Although these approaches demonstrate efficient and cost-effective segmentation performance in cross-domain implementations, they require additional procedures to put their striking characteristics into practical use. We use our method for developing a visual perception technique to perform in urban sidewalk environments for the robotic wheelchair. We generate a collection of synthetic scenes in a blending target distribution to train and validate our approach. Experimental results show that our method improves prediction accuracy on our benchmark with tolerable loss of speed and without additional overhead. Overall, our technique serves as a reference to transfer and develop perception algorithms for any cross-domain visual perception applications with less downtime. |
| first_indexed | 2024-03-09T10:12:07Z |
| format | Article |
| id | doaj.art-0cb4b1bb20a94e8c881afa5eea6b14d3 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T10:12:07Z |
| publishDate | 2022-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-0cb4b1bb20a94e8c881afa5eea6b14d32023-12-01T22:40:12ZengMDPI AGSensors1424-82202022-07-012214524110.3390/s22145241Self-Supervised Sidewalk Perception Using Fast Video Semantic Segmentation for Robotic Wheelchairs in Smart MobilityVishnu Pradeep0Redouane Khemmar1Louis Lecrosnier2Yann Duchemin3Romain Rossi4Benoit Decoux5Normandie University, UNIROUEN, ESIGELEC, IRSEEM, 76000 Rouen, FranceNormandie University, UNIROUEN, ESIGELEC, IRSEEM, 76000 Rouen, FranceNormandie University, UNIROUEN, ESIGELEC, IRSEEM, 76000 Rouen, FranceNormandie University, UNIROUEN, ESIGELEC, IRSEEM, 76000 Rouen, FranceNormandie University, UNIROUEN, ESIGELEC, IRSEEM, 76000 Rouen, FranceNormandie University, UNIROUEN, ESIGELEC, IRSEEM, 76000 Rouen, FranceThe real-time segmentation of sidewalk environments is critical to achieving autonomous navigation for robotic wheelchairs in urban territories. A robust and real-time video semantic segmentation offers an apt solution for advanced visual perception in such complex domains. The key to this proposition is to have a method with lightweight flow estimations and reliable feature extractions. We address this by selecting an approach based on recent trends in video segmentation. Although these approaches demonstrate efficient and cost-effective segmentation performance in cross-domain implementations, they require additional procedures to put their striking characteristics into practical use. We use our method for developing a visual perception technique to perform in urban sidewalk environments for the robotic wheelchair. We generate a collection of synthetic scenes in a blending target distribution to train and validate our approach. Experimental results show that our method improves prediction accuracy on our benchmark with tolerable loss of speed and without additional overhead. Overall, our technique serves as a reference to transfer and develop perception algorithms for any cross-domain visual perception applications with less downtime.https://www.mdpi.com/1424-8220/22/14/5241video semantic segmentationsidewalk segmentationcross-domainspatial convolutiondilated convolutionerror mitigation |
| spellingShingle | Vishnu Pradeep Redouane Khemmar Louis Lecrosnier Yann Duchemin Romain Rossi Benoit Decoux Self-Supervised Sidewalk Perception Using Fast Video Semantic Segmentation for Robotic Wheelchairs in Smart Mobility Sensors video semantic segmentation sidewalk segmentation cross-domain spatial convolution dilated convolution error mitigation |
| title | Self-Supervised Sidewalk Perception Using Fast Video Semantic Segmentation for Robotic Wheelchairs in Smart Mobility |
| title_full | Self-Supervised Sidewalk Perception Using Fast Video Semantic Segmentation for Robotic Wheelchairs in Smart Mobility |
| title_fullStr | Self-Supervised Sidewalk Perception Using Fast Video Semantic Segmentation for Robotic Wheelchairs in Smart Mobility |
| title_full_unstemmed | Self-Supervised Sidewalk Perception Using Fast Video Semantic Segmentation for Robotic Wheelchairs in Smart Mobility |
| title_short | Self-Supervised Sidewalk Perception Using Fast Video Semantic Segmentation for Robotic Wheelchairs in Smart Mobility |
| title_sort | self supervised sidewalk perception using fast video semantic segmentation for robotic wheelchairs in smart mobility |
| topic | video semantic segmentation sidewalk segmentation cross-domain spatial convolution dilated convolution error mitigation |
| url | https://www.mdpi.com/1424-8220/22/14/5241 |
| work_keys_str_mv | AT vishnupradeep selfsupervisedsidewalkperceptionusingfastvideosemanticsegmentationforroboticwheelchairsinsmartmobility AT redouanekhemmar selfsupervisedsidewalkperceptionusingfastvideosemanticsegmentationforroboticwheelchairsinsmartmobility AT louislecrosnier selfsupervisedsidewalkperceptionusingfastvideosemanticsegmentationforroboticwheelchairsinsmartmobility AT yannduchemin selfsupervisedsidewalkperceptionusingfastvideosemanticsegmentationforroboticwheelchairsinsmartmobility AT romainrossi selfsupervisedsidewalkperceptionusingfastvideosemanticsegmentationforroboticwheelchairsinsmartmobility AT benoitdecoux selfsupervisedsidewalkperceptionusingfastvideosemanticsegmentationforroboticwheelchairsinsmartmobility |