Investigasi Pengaruh Step Training pada Metode Single Shot Multibox Detector untuk Marker dalam Teknologi Augmented Reality
DOI:
https://doi.org/10.22441/fifo.2020.v12i1.001Keywords:
Augmented Reality, Object Detection, Single shot multibox detector, Convolutional neural network, Transfer LearningAbstract
Nowadays, Artificial Intelligence is one of the most developing technology, especially on Augmented Reality (AR). AR is a technology which connected between real world and virtual in a real time that allows user to interact directly and display it in 3D. AR technology has two methods, that are AR based on marker and AR based on markerless. However, AR based on marker need an object detection system which has high performance as an interaction tools between user and the device. Single shot multibox detector (SSD) is an object detection algorithm that has fast learning computation and good performance. This method is affected by some parameters like number of epoch, learning rate, batch size, step training, etc. However, to create a good system it took a long process such as taking dataset, labelling process, then training and testing models to gain the best performance. In this experiment, we analyze SSD method in AR technology using inception architecture as pre-trained Convolutional neural network (CNN), and then do transfer learning to minimize amount training time. The configuration that used is the number of step training. The result of this experiment gets the best accuracy in 70.17%. Then, the best performance is used as an object detection model for marker’s AR technology.
Abstrak
Saat ini, Artificial intelligence merupakan teknologi yang sedang berkembang pesat. Salah satunya adalah teknologi Augmented Reality (AR). AR adalah teknologi yang menggabungkan dunia nyata dengan virtual secara real-time dengan interaksi pengguna secara langsung dan menampilkannya dalam bentuk 3D. Teknologi AR ini memiliki dua metode yaitu dengan marker dan markerless. Dalam perkembangannya, AR berbasis marker membutuhkan sistem deteksi objek yang memiliki performa tinggi sebagai alat interaksi antara pengguna dengan perangkatnya. Single shot multibox detector (SSD) merupakan algoritma deteksi objek yang memiliki komputasi pembelajaran dan kinerja yang baik. Metode ini dipengaruhi oleh beberapa parameter seperti jumlah lapisan konvolusi, epoch, learning rate, jumlah batch, step training, dll. Namun, dalam mengimplementasikannya diperlukan proses yang cukup panjang seperti, pengambilan dataset, proses pelabelan, proses pelatihan menggunakan metode SSD, dan melakukan pengujian terhadap beberapa model untuk mencari perfomansi paling baik. Dalam percobaan ini, kami melakukan analisis terhadap metode SSD pada teknologi AR menggunakan arsitektur Inception sebagai pre-trained Convolutional neural network (CNN), kemudian dilakukan transfer learning untuk memperkecil jumlah kelas data pelatihan dan waktu pelatihan data. Konfigurasi yang digunakan berupa jumlah step pada pelatihan. Hasil dari penilitian ini menunjukan akurasi terbaik sebesar 70,17%. Kemudian, perfomansi terbaik digunakan sebagai model deteksi objek untuk marker pada teknologi AR.
Downloads
References
Billinghurst, M., Clark, A., & Lee, G. 2015. A Survey of Augmented Reality. Foundations and Trends in Human–Computer Interaction, 8(2-3), 73–272.
S. Sadhana Rao. 2010. Sixth sense technology. International Conference on Communication and Computational Intelligence (INCOCCI), Erode.
S. Ren, K. He, R. Girshick, and J. Sun. 2017. Faster r-cnn: towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis & Machine Intelligence.
J. Redmon, S. Divvala, R. Girshick, and A. Farhadi. 2016. You only look once: Unified, real-time object detection. Proceedings of the IEEE conference on computer vision and pattern recognition.
M. A. Afwani, E. Utami, and E. Pramono. 2017. Modifikasi Default-Boxes Pada Model SSD Untuk Meningkatkan Keakuratan Deteksi. Jurnal IT CIDA.
Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based learning applied to document recognition,” Proceedings of the IEEE.
S. A. Wibowo, H. Lee, E. K. Kim, and S. Kim. 2018. Collaborative learning based on convolutional features and correlation filter for visual tracking. International Journal of Control, Automation and Systems.
S. A. Wibowo, H. Lee, E. K. Kim, and S. Kim. 2017. Convolutional shallow features for performance improvement of histogram of oriented gradients in visual object tracking. Mathematical Problems in Engineering.
S. A. Wibowo, H. Lee, E. K. Kim, and S. Kim. 2017. Visual Tracking Based on Complementary Learners with Distractor Handling. Mathematical Problems in Engineering.
J. Wu. 2017. Introduction to convolutional neural networks. National Key Lab for Novel Software Technology. Nanjing University. China.
N. Sofia, 2018. Convolutional neural network. A Medium Corporation.
R. Darmadi. 2018. Mengenal Convolutional Layer Dan Pooling layer. Medium Corporation.
A. Yanuar. 2018. Fully-Connected Layer CNN dan Implementasinya. Universitas Gadjah Mada Menara Machine learning.
Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.-Y., & Berg, A. C. 2016. SSD: Single shot multibox detector. Lecture Notes in Computer Science, 21–37.
C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna. 2016. Rethinking the inception architecture for computer vision. Proceedings of the IEEE conference on computer vision and pattern recognition.
V. Rahmawan, D. Oktavian, and D. Alamsyah. 2017. Penerapan Algoritma Particle Filter pada Face Tracking.
Tian, P. 2015. A particle filter object tracking based on feature and location fusion. 6th IEEE International Conference on Software Engineering and Service Science (ICSESS).
Downloads
Additional Files
Published
How to Cite
Issue
Section
License
The copyright to this article is transferred to Universitas Mercu Buana (UMB) if and when the article is accepted for publication. The undersigned hereby transfers any and all rights in and to the paper including without limitation all copyrights to UMB. The undersigned hereby represents and warrants that the paper is original and that he/she is the author of the paper, except for material that is clearly identified as to its original source, with permission notices from the copyright owners where required. The undersigned represents that he/she has the power and authority to make and execute this assignment.
We declare that this paper has not been published in the same form elsewhere.
Furthermore, I/We hereby transfer the unlimited rights of publication of the above-mentioned paper as a whole to UMB. The copyright transfer covers the right to reproduce and distribute the article, including reprints, translations, photographic reproductions, microform, electronic form (offline, online) or any other reproductions of similar nature.
The corresponding author signs for and accepts responsibility for releasing this material on behalf of any and all co-authors. This agreement is to be signed by at least one of the authors who have obtained the assent of the co-author(s) where applicable. After submission of this agreement signed by the corresponding author, changes of authorship or in the order of the authors listed will not be accepted.
Retained Rights/Terms and Conditions
Although authors are permitted to re-use all or portions of the Work in other works, this does not include granting third-party requests for reprinting, republishing, or other types of re-use.
Our Articles are licensed under CC BY-NC
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
