Introduction
Sketch a Mango Web Application is an online technology demonstration of Image-to-Image translation with Deep Convolutional Generative Advesarial Network.
One of the main use of this web application is to allow user to generate synthetic images of Sala Mango just by sketching the shape of the mango. The user also can add bruises to the mango just by highlighting the defect area on the sketch and the web application will generate realistic skin bruises on the mango image.
This web application could be used on any HTML5 compliant web browser, although for best experience, it is recommended to run it on a personal computer.
Sketch a Mango Web Application is trained using TensorFlow with Keras, and deployed with the help of TensorFlow.js library.
How to use this web application
This short video demonstrate how to use this web application
What is Sala Mango ?
Sala Mango (Mangifera Indica Linn) is a mango cultivar (Clone no: MA 249) grown in Malaysia. Also known as "Perlis Sunshine" as it is popularly cultivated in Perlis. This cultivars is known to have grenish skin color when ripe and yellowish flesh with a distintive sweet taste. Sala Mango cultivars is known to have high resiliency againts diseases and pest attack compared to other mango cultivars grown in the northern sates in Malaysia.
Dataset used in this web application
The dataset used in this application is Sala Mango Images dataset, and may be cited as:
Mohammad Hafiz bin Ismail. (2022). Sala Mango Images Dataset [Data set]. Kaggle. https://doi.org/10.34740/KAGGLE/DSV/3183377
Author
Sketch a Mango Web Application is authored by :
Mohammad Hafiz bin Ismail <mypapit@gmail.com>.
For any enquiries, please send an email to deeplearn@uitm.edu.my
Referencing this work
You can download a BibTex reference file or you may refer to this work as:
Ismail, M. H. (2022). Sketch a Mango Image-to-Image GAN Web Application - release 1.0. Retrieved , , from: https://demo.mobilepit.com/ai/drawmango. doi:10.5281/zenodo.6562805
Further Readings
If you are curious about GAN, Image to Image translation and other AI techniques behind this Web Application and further applications, you may want to read these academic papers:
- Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., ... & Bengio, Y. (2014). Generative adversarial nets. Advances in neural information processing systems, 27.
- Isola, Phillip, et al. Image-to-image translation with conditional adversarial networks." Proceedings of the IEEE conference on computer vision and pattern recognition. 2017.
- Lin, J., Xia, Y., Qin, T., Chen, Z., & Liu, T. Y. (2018). Conditional image-to-image translation. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 5524-5532).
- Solanki, A., Nayyar, A., & Naved, M. (Eds.). (2021). Generative Adversarial Networks for Image-to-Image Translation. Academic Press.