Semantic web-mining and deep vision for lifelong object discovery
Autonomous robots that are to assist humans in their daily lives must recognize and understand the meaning of objects in their environment. However, the open nature of the world means robots must be able to learn and extend their knowledge about previously unknown objects on-line. In this work we in...
| Main Authors: | , , , , , |
|---|---|
| Format: | Conference item |
| Published: |
IEEE
2017
|
| _version_ | 1826301573353963520 |
|---|---|
| author | Young, J Kunze, L Basile, V Cabrio, E Hawes, N Caputo, B |
| author_facet | Young, J Kunze, L Basile, V Cabrio, E Hawes, N Caputo, B |
| author_sort | Young, J |
| collection | OXFORD |
| description | Autonomous robots that are to assist humans in their daily lives must recognize and understand the meaning of objects in their environment. However, the open nature of the world means robots must be able to learn and extend their knowledge about previously unknown objects on-line. In this work we investigate the problem of unknown object hypotheses generation, and employ a semantic Web-mining framework along with deep-learning-based object detectors. This allows us to make use of both visual and semantic features in combined hypotheses generation. Experiments on data from mobile robots in real world application deployments show that this combination improves performance over the use of either method in isolation. |
| first_indexed | 2024-03-07T05:34:27Z |
| format | Conference item |
| id | oxford-uuid:e366f854-c67a-443b-bf07-1a8e5b7d1fcb |
| institution | University of Oxford |
| last_indexed | 2024-03-07T05:34:27Z |
| publishDate | 2017 |
| publisher | IEEE |
| record_format | dspace |
| spelling | oxford-uuid:e366f854-c67a-443b-bf07-1a8e5b7d1fcb2022-03-27T10:08:51ZSemantic web-mining and deep vision for lifelong object discoveryConference itemhttp://purl.org/coar/resource_type/c_5794uuid:e366f854-c67a-443b-bf07-1a8e5b7d1fcbSymplectic Elements at OxfordIEEE2017Young, JKunze, LBasile, VCabrio, EHawes, NCaputo, BAutonomous robots that are to assist humans in their daily lives must recognize and understand the meaning of objects in their environment. However, the open nature of the world means robots must be able to learn and extend their knowledge about previously unknown objects on-line. In this work we investigate the problem of unknown object hypotheses generation, and employ a semantic Web-mining framework along with deep-learning-based object detectors. This allows us to make use of both visual and semantic features in combined hypotheses generation. Experiments on data from mobile robots in real world application deployments show that this combination improves performance over the use of either method in isolation. |
| spellingShingle | Young, J Kunze, L Basile, V Cabrio, E Hawes, N Caputo, B Semantic web-mining and deep vision for lifelong object discovery |
| title | Semantic web-mining and deep vision for lifelong object discovery |
| title_full | Semantic web-mining and deep vision for lifelong object discovery |
| title_fullStr | Semantic web-mining and deep vision for lifelong object discovery |
| title_full_unstemmed | Semantic web-mining and deep vision for lifelong object discovery |
| title_short | Semantic web-mining and deep vision for lifelong object discovery |
| title_sort | semantic web mining and deep vision for lifelong object discovery |
| work_keys_str_mv | AT youngj semanticwebmininganddeepvisionforlifelongobjectdiscovery AT kunzel semanticwebmininganddeepvisionforlifelongobjectdiscovery AT basilev semanticwebmininganddeepvisionforlifelongobjectdiscovery AT cabrioe semanticwebmininganddeepvisionforlifelongobjectdiscovery AT hawesn semanticwebmininganddeepvisionforlifelongobjectdiscovery AT caputob semanticwebmininganddeepvisionforlifelongobjectdiscovery |