Upper body detection and tracking in extended signing sequences
<p>The goal of this work is to detect and track the articulated pose of a human in signing videos of more than one hour in length. In particular we wish to accurately localise hands and arms, despite fast motion and a cluttered and changing background.</p><p>We cast the problem as...
| Main Authors: | , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Springer
2011
|
| Subjects: |
| _version_ | 1811140710831751168 |
|---|---|
| author | Buehler, P Everingham, M Huttenlocher, D Zisserman, A |
| author_facet | Buehler, P Everingham, M Huttenlocher, D Zisserman, A |
| author_sort | Buehler, P |
| collection | OXFORD |
| description | <p>The goal of this work is to detect and track the articulated pose of a human in signing videos of more than one hour in length. In particular we wish to accurately localise hands and arms, despite fast motion and a cluttered and changing background.</p><p>We cast the problem as inference in a generative model of the image, and propose a complete model which accounts for self-occlusion of the arms. Under this model, limb detection is expensive due to the very large number of possible configurations each part can assume. We make the following contributions to reduce this cost: (i) efficient sampling from a pictorial structure proposal distribution to obtain reasonable configurations; (ii) identifying a large number of frames where configurations can be correctly inferred, and exploiting temporal tracking elsewhere.</p><p>Results are reported for signing footage with challenging image conditions and for different signers. We show that the method is able to identify the true arm and hand locations with high reliability. The results exceed the state-of-the-art for the length and stability of continuous limb tracking.</p> |
| first_indexed | 2024-03-06T21:20:50Z |
| format | Journal article |
| id | oxford-uuid:41643af0-bb8b-4f18-a696-22eb623ce70c |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-09-25T04:26:19Z |
| publishDate | 2011 |
| publisher | Springer |
| record_format | dspace |
| spelling | oxford-uuid:41643af0-bb8b-4f18-a696-22eb623ce70c2024-08-23T10:29:49ZUpper body detection and tracking in extended signing sequencesJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:41643af0-bb8b-4f18-a696-22eb623ce70cEngineering & allied sciencesEnglishOxford University Research Archive - ValetSpringer2011Buehler, PEveringham, MHuttenlocher, DZisserman, A<p>The goal of this work is to detect and track the articulated pose of a human in signing videos of more than one hour in length. In particular we wish to accurately localise hands and arms, despite fast motion and a cluttered and changing background.</p><p>We cast the problem as inference in a generative model of the image, and propose a complete model which accounts for self-occlusion of the arms. Under this model, limb detection is expensive due to the very large number of possible configurations each part can assume. We make the following contributions to reduce this cost: (i) efficient sampling from a pictorial structure proposal distribution to obtain reasonable configurations; (ii) identifying a large number of frames where configurations can be correctly inferred, and exploiting temporal tracking elsewhere.</p><p>Results are reported for signing footage with challenging image conditions and for different signers. We show that the method is able to identify the true arm and hand locations with high reliability. The results exceed the state-of-the-art for the length and stability of continuous limb tracking.</p> |
| spellingShingle | Engineering & allied sciences Buehler, P Everingham, M Huttenlocher, D Zisserman, A Upper body detection and tracking in extended signing sequences |
| title | Upper body detection and tracking in extended signing sequences |
| title_full | Upper body detection and tracking in extended signing sequences |
| title_fullStr | Upper body detection and tracking in extended signing sequences |
| title_full_unstemmed | Upper body detection and tracking in extended signing sequences |
| title_short | Upper body detection and tracking in extended signing sequences |
| title_sort | upper body detection and tracking in extended signing sequences |
| topic | Engineering & allied sciences |
| work_keys_str_mv | AT buehlerp upperbodydetectionandtrackinginextendedsigningsequences AT everinghamm upperbodydetectionandtrackinginextendedsigningsequences AT huttenlocherd upperbodydetectionandtrackinginextendedsigningsequences AT zissermana upperbodydetectionandtrackinginextendedsigningsequences |