Double Difference Motion Detection and Its Application for Madura Batik Virtual Fitting Room
Abstract: Madura Batik Virtual
Fitting Room using double difference algorithms motion detection is proposed in
this research. This virtual fitting room consists of three main stages, i.e.
motion detection, determination of the region of interest of the detected
motion, superimposed the virtual clothes into the region of interest. The
double difference algorithm is used for the motion detection stage, since in
this algorithm, the empty frame as the reference frame is not required. The
double difference algorithm uses the previous and next frame to detect the
motion in the current frame. Perception Test Images Sequences Dataset are used
as the data of the experiment to measure the performance accuracy of this algorithm
before the algorithm is used for the Madura batik virtual fitting room. The
accuracy is 57.31%, 99.71%, and 78.52% for the sensitivity, specificity, and
balanced accuracy, respectively. The build Madura batik virtual fitting room in
this research can be used as the added feature of the Madura batik online
stores, hence the consumer is able to see whether the clothes is fitted to them
or not, and this virtual fitting room is also can be used as the promotion of
Madura batik broadly.
Author: Rima Triwahyuningrum,
Indah Agustien Siradjuddin, Yonathan Fery Hendrawan,
Journal Code: jptkomputergg150172