Find your own way: Weakly-supervised segmentation of path proposals for urban autonomy
We present a weakly-supervised approach to segmenting proposed drivable paths in images with the goal of autonomous driving in complex urban environments. Using recorded routes from a data collection vehicle, our proposed method generates vast quantities of labelled images containing proposed paths...
| Main Authors: | , , |
|---|---|
| 格式: | Conference item |
| 出版: |
Institute of Electrical and Electronics Engineers
2017
|
| _version_ | 1826271437200031744 |
|---|---|
| author | Barnes, D Maddern, W Posner, H |
| author_facet | Barnes, D Maddern, W Posner, H |
| author_sort | Barnes, D |
| collection | OXFORD |
| description | We present a weakly-supervised approach to segmenting proposed drivable paths in images with the goal of autonomous driving in complex urban environments. Using recorded routes from a data collection vehicle, our proposed method generates vast quantities of labelled images containing proposed paths and obstacles without requiring manual annotation, which we then use to train a deep semantic segmentation network. With the trained network we can segment proposed paths and obstacles at run-time using a vehicle equipped with only a monocular camera without relying on explicit modelling of road or lane markings. We evaluate our method on the large-scale KITTI and Oxford RobotCar datasets and demonstrate reliable path proposal and obstacle segmentation in a wide variety of environments under a range of lighting, weather and traffic conditions. We illustrate how the method can generalise to multiple path proposals at intersections and outline plans to incorporate the system into a framework for autonomous urban driving. |
| first_indexed | 2024-03-06T21:56:39Z |
| format | Conference item |
| id | oxford-uuid:4d24f0d3-c9cc-4452-a077-e3d38efff41f |
| institution | University of Oxford |
| last_indexed | 2024-03-06T21:56:39Z |
| publishDate | 2017 |
| publisher | Institute of Electrical and Electronics Engineers |
| record_format | dspace |
| spelling | oxford-uuid:4d24f0d3-c9cc-4452-a077-e3d38efff41f2022-03-26T15:53:45ZFind your own way: Weakly-supervised segmentation of path proposals for urban autonomyConference itemhttp://purl.org/coar/resource_type/c_5794uuid:4d24f0d3-c9cc-4452-a077-e3d38efff41fSymplectic Elements at OxfordInstitute of Electrical and Electronics Engineers2017Barnes, DMaddern, WPosner, HWe present a weakly-supervised approach to segmenting proposed drivable paths in images with the goal of autonomous driving in complex urban environments. Using recorded routes from a data collection vehicle, our proposed method generates vast quantities of labelled images containing proposed paths and obstacles without requiring manual annotation, which we then use to train a deep semantic segmentation network. With the trained network we can segment proposed paths and obstacles at run-time using a vehicle equipped with only a monocular camera without relying on explicit modelling of road or lane markings. We evaluate our method on the large-scale KITTI and Oxford RobotCar datasets and demonstrate reliable path proposal and obstacle segmentation in a wide variety of environments under a range of lighting, weather and traffic conditions. We illustrate how the method can generalise to multiple path proposals at intersections and outline plans to incorporate the system into a framework for autonomous urban driving. |
| spellingShingle | Barnes, D Maddern, W Posner, H Find your own way: Weakly-supervised segmentation of path proposals for urban autonomy |
| title | Find your own way: Weakly-supervised segmentation of path proposals for urban autonomy |
| title_full | Find your own way: Weakly-supervised segmentation of path proposals for urban autonomy |
| title_fullStr | Find your own way: Weakly-supervised segmentation of path proposals for urban autonomy |
| title_full_unstemmed | Find your own way: Weakly-supervised segmentation of path proposals for urban autonomy |
| title_short | Find your own way: Weakly-supervised segmentation of path proposals for urban autonomy |
| title_sort | find your own way weakly supervised segmentation of path proposals for urban autonomy |
| work_keys_str_mv | AT barnesd findyourownwayweaklysupervisedsegmentationofpathproposalsforurbanautonomy AT maddernw findyourownwayweaklysupervisedsegmentationofpathproposalsforurbanautonomy AT posnerh findyourownwayweaklysupervisedsegmentationofpathproposalsforurbanautonomy |