dc.contributor.author | Kuchar, Olga Anna. | en_US |
dc.date.accessioned | 2014-10-21T12:33:19Z | |
dc.date.available | 1999 | |
dc.date.issued | 1999 | en_US |
dc.identifier.other | AAINQ39322 | en_US |
dc.identifier.uri | http://hdl.handle.net/10222/55615 | |
dc.description | Adding virtual humans into virtual reality environments requires expensive hardware and, for the animator, time-consuming tasks. An ideal situation in populating virtual worlds would be to create an artificially intelligent virtual human (AIVH) that is capable of moving and interacting in its environment with little intervention from the animator. This research focuses on the preliminary developments of creating such an AIVH. Before an AIVH can move completely, an AIVH needs to learn how to move its body. Thus, the body needs to be divided into separate areas and the AIVH needs to learn about movements in these areas. | en_US |
dc.description | This thesis focuses on free hand motion, and more specifically on thumb and finger postures. The author incorporates the use of feedforward artificial neural networks (ANNs) to manipulate an underlying model in real time and determine postures of the thumb and fingers to allow for dynamic hand animation. The underlying model of the computerized human hand is based on a biological model involving tendons. The ANNs trigger any tendons that need to be manipulated to achieve an animated goal. The prescribed animated goals involve different flexion, extension, abduction, and adduction movements. There are several drawbacks associated with using a multilayer feedforward network. First, since the architecture is designed by trial-and-error, it is very difficult to create an effective 'a priori' architecture. Secondly, ANNs are not trained on an entire input space, but there is a supposition that the ANN will work correctly when presented with any possible input within this problem space. | en_US |
dc.description | In this ongoing research, the ability for ANNs to determine tendon control in the thumb and fingers has been developed and tested, producing positive results. Now that ANNs are seen to be successfully applied to hands, the next step is to apply this technique to other parts of the body. | en_US |
dc.description | Thesis (Ph.D.)--DalTech - Dalhousie University (Canada), 1999. | en_US |
dc.language | eng | en_US |
dc.publisher | Dalhousie University | en_US |
dc.publisher | | en_US |
dc.subject | Engineering, Electronics and Electrical. | en_US |
dc.subject | Artificial Intelligence. | en_US |
dc.subject | Computer Science. | en_US |
dc.title | Development of animated finger movements via a neural network for tendon tension control. | en_US |
dc.type | text | en_US |
dc.contributor.degree | Ph.D. | en_US |