{"id":{"repo_id":"southwales","oai_identifier":"oai:pure.atira.dk:studenttheses/f375e712-038c-4a78-862a-944c0e36e360"},"canonical_url":"https://search.dev.ndltd.org/etd/southwales/oai:pure.atira.dk:studenttheses/f375e712-038c-4a78-862a-944c0e36e360","repository":{"repo_id":"southwales","name":"University of South Wales","base_url":"https://pure.southwales.ac.uk/ws/oai"},"display":{"title":"Evolved neural network approximation of discontinuous vector fields in unit quaternion space (S3) for anatomical joint constraint","abstract":"The creation of anatomically correct three-dimensional joints for the simulation of humans is a complex process, a key difficulty being the correction of invalid joint configurations to the nearest valid alternative. Personalised models based on individual joint mobility are in demand in both animation and medicine [1]. Medical models need to be highly accurate animated models less so, however if either are to be used in a real time environment they must have a low temporal cost (high performance). This work briefly explores Support Vector Machine neural networks as joint configuration classifiers that group joint configurations into invalid and valid. A far more detailed investigation is carried out into the use of topologically evolved feed forward neural networks for the generation of appropriately proportioned corrective components which when applied to an invalid joint configuration result in a valid configuration and the same configuration if the original configuration was valid. Discontinuous vector fields<br/>were used to represent constraints of varying size, dimensionality and complexity. This culminated in the creation corrective quaternion constraints represented by discontinuous vector fields, learned by topologically evolved neural networks and trained via the resilient back propagation algorithm. Quaternion constraints are difficult to implement and although alternative methods exist [2-6] the method presented here is superior in many respects. This method of joint constraint forms the basis of the contribution to knowledge along with the discovery of relationships between the continuity and distribution of samples in quaternion space and neural network performance. The results of the experiments for constraints on the rotation of limb with regular boundaries show that 3.7 x 10<sup>-4</sup>% of patterns resulted in errors greater than 2% of the maximum possible error while for irregular boundaries 0.032% of patterns resulted in errors greater than 7.5%.","abstract_html":"The creation of anatomically correct three-dimensional joints for the simulation of humans is a complex process, a key difficulty being the correction of invalid joint configurations to the nearest valid alternative. Personalised models based on individual joint mobility are in demand in both animation and medicine [1]. Medical models need to be highly accurate animated models less so, however if either are to be used in a real time environment they must have a low temporal cost (high performance). This work briefly explores Support Vector Machine neural networks as joint configuration classifiers that group joint configurations into invalid and valid. A far more detailed investigation is carried out into the use of topologically evolved feed forward neural networks for the generation of appropriately proportioned corrective components which when applied to an invalid joint configuration result in a valid configuration and the same configuration if the original configuration was valid. Discontinuous vector fields&lt;br/&gt;were used to represent constraints of varying size, dimensionality and complexity. This culminated in the creation corrective quaternion constraints represented by discontinuous vector fields, learned by topologically evolved neural networks and trained via the resilient back propagation algorithm. Quaternion constraints are difficult to implement and although alternative methods exist [2-6] the method presented here is superior in many respects. This method of joint constraint forms the basis of the contribution to knowledge along with the discovery of relationships between the continuity and distribution of samples in quaternion space and neural network performance. The results of the experiments for constraints on the rotation of limb with regular boundaries show that 3.7 x 10&lt;sup&gt;-4&lt;/sup&gt;% of patterns resulted in errors greater than 2% of the maximum possible error while for irregular boundaries 0.032% of patterns resulted in errors greater than 7.5%.","abstract_has_math":false,"creators":["Jenkins, Glenn"],"institution":null,"degree_name":"Doctoral Thesis","degree_level":"Student thesis","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Angel, Paul"],"committee_chairs":[],"committee_members":[],"year":2007,"date_issued":"2007-8","date_published":"2007-8","updated_at":"2026-07-24T04:38:46Z","subjects":["Human locomotion","Neural networks (Computer science)"],"languages":["eng"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:pure.atira.dk:studenttheses/f375e712-038c-4a78-862a-944c0e36e360"],"render_values":[{"text":"oai:pure.atira.dk:studenttheses/f375e712-038c-4a78-862a-944c0e36e360","href":null,"code":true}]}]},"links":{"outbound_url":"https://pure.southwales.ac.uk/en/studentTheses/f375e712-038c-4a78-862a-944c0e36e360","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Angel, Paul"]},{"key":"dc:creator","label":"Author","values":["Jenkins, Glenn"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2007-8"]},{"key":"dc:date.issued","label":"Date","values":["2007-8"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://pure.southwales.ac.uk/en/studentTheses/f375e712-038c-4a78-862a-944c0e36e360"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Student thesis"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["Doctoral Thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Human locomotion","Neural networks (Computer science)"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["oai:pure.atira.dk:studenttheses/f375e712-038c-4a78-862a-944c0e36e360","https://pure.southwales.ac.uk/en/studentTheses/f375e712-038c-4a78-862a-944c0e36e360"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://pure.southwales.ac.uk/files/1907591/G._L._Jenkins_2008_2060371.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The creation of anatomically correct three-dimensional joints for the simulation of humans is a complex process, a key difficulty being the correction of invalid joint configurations to the nearest valid alternative. Personalised models based on individual joint mobility are in demand in both animation and medicine [1]. Medical models need to be highly accurate animated models less so, however if either are to be used in a real time environment they must have a low temporal cost (high performance). This work briefly explores Support Vector Machine neural networks as joint configuration classifiers that group joint configurations into invalid and valid. A far more detailed investigation is carried out into the use of topologically evolved feed forward neural networks for the generation of appropriately proportioned corrective components which when applied to an invalid joint configuration result in a valid configuration and the same configuration if the original configuration was valid. Discontinuous vector fields<br/>were used to represent constraints of varying size, dimensionality and complexity. This culminated in the creation corrective quaternion constraints represented by discontinuous vector fields, learned by topologically evolved neural networks and trained via the resilient back propagation algorithm. Quaternion constraints are difficult to implement and although alternative methods exist [2-6] the method presented here is superior in many respects. This method of joint constraint forms the basis of the contribution to knowledge along with the discovery of relationships between the continuity and distribution of samples in quaternion space and neural network performance. The results of the experiments for constraints on the rotation of limb with regular boundaries show that 3.7 x 10<sup>-4</sup>% of patterns resulted in errors greater than 2% of the maximum possible error while for irregular boundaries 0.032% of patterns resulted in errors greater than 7.5%."]},{"key":"dc:title","label":"Title","values":["Evolved neural network approximation of discontinuous vector fields in unit quaternion space (S3) for anatomical joint constraint"]}]}],"canonical_facts":{"dc:contributor.advisor":["Angel, Paul"],"dc:creator":["Jenkins, Glenn"],"dc:date":["2007-8"],"dc:date.issued":["2007-8"],"dc:description.abstract":["The creation of anatomically correct three-dimensional joints for the simulation of humans is a complex process, a key difficulty being the correction of invalid joint configurations to the nearest valid alternative. Personalised models based on individual joint mobility are in demand in both animation and medicine [1]. Medical models need to be highly accurate animated models less so, however if either are to be used in a real time environment they must have a low temporal cost (high performance). This work briefly explores Support Vector Machine neural networks as joint configuration classifiers that group joint configurations into invalid and valid. A far more detailed investigation is carried out into the use of topologically evolved feed forward neural networks for the generation of appropriately proportioned corrective components which when applied to an invalid joint configuration result in a valid configuration and the same configuration if the original configuration was valid. Discontinuous vector fields<br/>were used to represent constraints of varying size, dimensionality and complexity. This culminated in the creation corrective quaternion constraints represented by discontinuous vector fields, learned by topologically evolved neural networks and trained via the resilient back propagation algorithm. Quaternion constraints are difficult to implement and although alternative methods exist [2-6] the method presented here is superior in many respects. This method of joint constraint forms the basis of the contribution to knowledge along with the discovery of relationships between the continuity and distribution of samples in quaternion space and neural network performance. The results of the experiments for constraints on the rotation of limb with regular boundaries show that 3.7 x 10<sup>-4</sup>% of patterns resulted in errors greater than 2% of the maximum possible error while for irregular boundaries 0.032% of patterns resulted in errors greater than 7.5%."],"dc:identifier":["oai:pure.atira.dk:studenttheses/f375e712-038c-4a78-862a-944c0e36e360","https://pure.southwales.ac.uk/en/studentTheses/f375e712-038c-4a78-862a-944c0e36e360"],"dc:identifier.uri":["https://pure.southwales.ac.uk/files/1907591/G._L._Jenkins_2008_2060371.pdf"],"dc:language":["eng"],"dc:relation.isreferencedby":["https://pure.southwales.ac.uk/en/studentTheses/f375e712-038c-4a78-862a-944c0e36e360"],"dc:subject":["Human locomotion","Neural networks (Computer science)"],"dc:title":["Evolved neural network approximation of discontinuous vector fields in unit quaternion space (S3) for anatomical joint constraint"],"dc:type":["Thesis"],"dc:type.qualificationlevel":["Student thesis"],"dc:type.qualificationname":["Doctoral Thesis"]},"updated_at":"2026-07-24T04:38:46Z"}