Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques
In this paper, the notion of risk analysis within 3D scenes using vision based techniques is introduced. In particular the problem of risk estimation of indoor environments at the scene and object level is considered, with applications in domestic robots and smart homes. To this end, the proposed Ri...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Published: |
Elsevier
2016
|
| Subjects: | |
| Online Access: | https://eprints.nottingham.ac.uk/37237/ |
| _version_ | 1848795419542290432 |
|---|---|
| author | Dupre, Rob Argyriou, Vasileios Tzimiropoulos, Georgios Greenhill, Darrel |
| author_facet | Dupre, Rob Argyriou, Vasileios Tzimiropoulos, Georgios Greenhill, Darrel |
| author_sort | Dupre, Rob |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | In this paper, the notion of risk analysis within 3D scenes using vision based techniques is introduced. In particular the problem of risk estimation of indoor environments at the scene and object level is considered, with applications in domestic robots and smart homes. To this end, the proposed Risk Estimation Framework is described, which provides a quantified risk score for a given scene. This methodology is extended with the introduction of a novel robust kernel for 3D shape descriptors such as 3D HOG and SIFT3D, which aims to reduce the effects of outliers in the proposed risk recognition methodology. The Physics Behaviour Feature (PBF) is presented, which uses an object's angular velocity obtained using Newtonian physics simulation as a descriptor. Furthermore, an extension of boosting techniques for learning is suggested in the form of the novel Complex and Hyper-Complex Adaboost, which greatly increase the computation efficiency of the original technique. In order to evaluate the proposed robust descriptors an enriched version of the 3D Risk Scenes (3DRS) dataset with extra objects, scenes and meta-data was utilised. A comparative study was conducted demonstrating that the suggested approach outperforms current state-of-the-art descriptors. |
| first_indexed | 2025-11-14T19:31:47Z |
| format | Article |
| id | nottingham-37237 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| last_indexed | 2025-11-14T19:31:47Z |
| publishDate | 2016 |
| publisher | Elsevier |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-372372020-05-04T18:18:37Z https://eprints.nottingham.ac.uk/37237/ Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques Dupre, Rob Argyriou, Vasileios Tzimiropoulos, Georgios Greenhill, Darrel In this paper, the notion of risk analysis within 3D scenes using vision based techniques is introduced. In particular the problem of risk estimation of indoor environments at the scene and object level is considered, with applications in domestic robots and smart homes. To this end, the proposed Risk Estimation Framework is described, which provides a quantified risk score for a given scene. This methodology is extended with the introduction of a novel robust kernel for 3D shape descriptors such as 3D HOG and SIFT3D, which aims to reduce the effects of outliers in the proposed risk recognition methodology. The Physics Behaviour Feature (PBF) is presented, which uses an object's angular velocity obtained using Newtonian physics simulation as a descriptor. Furthermore, an extension of boosting techniques for learning is suggested in the form of the novel Complex and Hyper-Complex Adaboost, which greatly increase the computation efficiency of the original technique. In order to evaluate the proposed robust descriptors an enriched version of the 3D Risk Scenes (3DRS) dataset with extra objects, scenes and meta-data was utilised. A comparative study was conducted demonstrating that the suggested approach outperforms current state-of-the-art descriptors. Elsevier 2016-12-01 Article PeerReviewed Dupre, Rob, Argyriou, Vasileios, Tzimiropoulos, Georgios and Greenhill, Darrel (2016) Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques. Information Sciences, 372 . pp. 359-379. ISSN 1872-6291 3D Scene analysis Risk Estimation Domestic robots Smart homes HOG 3D VHOG http://www.sciencedirect.com/science/article/pii/S0020025516306570 doi:10.1016/j.ins.2016.08.075 doi:10.1016/j.ins.2016.08.075 |
| spellingShingle | 3D Scene analysis Risk Estimation Domestic robots Smart homes HOG 3D VHOG Dupre, Rob Argyriou, Vasileios Tzimiropoulos, Georgios Greenhill, Darrel Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques |
| title | Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques |
| title_full | Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques |
| title_fullStr | Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques |
| title_full_unstemmed | Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques |
| title_short | Risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques |
| title_sort | risk analysis for smart homes and domestic robots using robust shape and physics descriptors, and complex boosting techniques |
| topic | 3D Scene analysis Risk Estimation Domestic robots Smart homes HOG 3D VHOG |
| url | https://eprints.nottingham.ac.uk/37237/ https://eprints.nottingham.ac.uk/37237/ https://eprints.nottingham.ac.uk/37237/ |