DESIGN AND IMPLEMENTATION HAND MOVEMENT AND BODY DETECTION (CASE STUDIES IN FASHION SHOPPING ASSISTANCE)
Design and implementation detection of hand movements and body in an online application for sale of clothing called Fashion Shopping Assistance (FSA). The appication use technology that combine hand motion detection for creating a virtual menu as an input to the computer. With using virtual menu use...
Saved in:
Main Author: | |
---|---|
Format: | Theses |
Language: | Indonesia |
Online Access: | https://digilib.itb.ac.id/gdl/view/15774 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Institut Teknologi Bandung |
Language: | Indonesia |
Summary: | Design and implementation detection of hand movements and body in an online application for sale of clothing called Fashion Shopping Assistance (FSA). The appication use technology that combine hand motion detection for creating a virtual menu as an input to the computer. With using virtual menu users can perform some activities such as changing clothing model, stores images, etc. Detection of the body mixed with Augmented Reality (AR) using the marker developed to provide a new user experience in buying or selling clothes and virtually try clothing. In this thesis use only one still camera detector and one marker as input to the computer. As input to computer the virtual menu are build with combination of blur filter, blend and the threshold so that only hand movements were detected and create a motion areas. Detection of the body is based on face detection for creating body shape modeling. Combination between body shape modeling and 3D models are expected to provide a realistic impression of clothes. FSA applications capable to detect movement of hand that build form a motion area required by a virtual menu. Virtual menu has not affected influence of environmental factors, the chance of success of the detection of at least 0,8. Face detection use front face, that can be used to form the body shape modeling, Combination and 3D models AR when the user's body moves. Error for attachment body shape modeling and AR are obtained at least 16 %. |
---|