In addition, please note that the deadline for submission of works in this competition is December 2nd, all developers must submit their works before then!

Go to the official introduction page of the Contest:

Developer. The android. Google. Cn/dev – challen…

  • Tencent Video Link

V.qq.com/x/page/e302…

  • Bilibili video link

www.bilibili.com/video/av769…

Enabling technology new Agriculture: Using machine learning to control insect infestation

Click the link below to view the video:

V.qq.com/x/page/m302…

Nazirini Siraji and several other developers noticed that farmers in Uganda were suffering from pests and decided to teach themselves TensorFlow to help local farmers. They took pictures of nearby farmland and collected training data. Then they retrained the transfer learning model MobileNet using TensorFlow and generated a TensorFlow Lite FlatBuffer file using the TensorFlow converter. Finally, deploy the file into the Android application. After farmers take and import photos of their crops, the app can analyze one frame of the picture to determine whether the crop is currently suffering from fall armyworm infestation, and then provide possible solutions based on the results. How about that? Isn’t that a great app?

Visit TensorFlow’s official website:

tensorflow.google.cn

Empowering the healthcare industry: Using machine learning to identify respiratory diseases

Click the link below to view the video:

V.qq.com/x/page/n302…

Tambua Health has developed a medical app using TensorFlow Lite that turns a smartphone into a powerful, non-invasive screening tool to help doctors diagnose patients with respiratory disease. The app helps doctors analyze and identify any unusual breathing sounds in the lungs, such as rales, crackles and wheezes.

Empowering Retail: Adidas uses machine learning to build superior shopping experiences

Machine learning is infusing innovation into industries, and retail is no exception. Adidas recently launched an app for its Flagship Store in London called Bring It to Me, which uses machine learning to provide consumers with an interactive shopping experience. Customers can use the Visual Lookup (image search) function in the app to scan the items in the store and check the inventory of the selected style as they walk through the store, without waiting in line to select items of their own size. Behind the great experience is technical support from ML Kit. This tech-retail encounter explores the practical use cases of machine learning from a new perspective and shows us how well machine learning can solve problems.

Advantages of machine learning based on terminal devices

Machine learning directly on the device has many benefits. First, the amount of data transferred to the server is significantly reduced, resulting in a significant increase in user privacy. Second, because the model runs entirely locally, it can work even offline, which is ideal for scenarios where there is no signal, such as rainforest hinterlands, deserts or the London Underground. Best of all, running the model locally on the device enables ultra-low latency, allowing you to unlock a wide variety of new user experiences. Machine learning is not only about process automation, but also about empowering people to solve problems better.

Google also provides rich technical support to help developers make full use of machine learning technology to create more exciting:

  • ML Kit provides developers with a one-stop SDK suite to help you solve problems using Google’s powerful machine learning model.
  • The TensorFlow Lite framework allows you to create custom models and optimize them for running on the Android platform.
  • Firebase and Google Cloud are two infrastructure platforms that can help you train your models. You can train specific types of native models with AutoML Vision Edge, or train your own models with the computing power of the platform.

If you have a novel idea you want to implement and want to help people solve real problems, please do it! We will select 10 finalists and provide professional guidance to help you develop a plan to make your ideas a real force in improving people’s lives! When the app is developed, we will show it to hundreds of millions of users in multiple forms on Google Play. Please go to the contest website to see what wonderful prizes we have prepared for you.

We will close the channel for proposal submission on December 2nd, please be sure to submit your proposal before then, we are looking forward to your wonderful performance! We hope to work with you to make use of the machine learning technology built into the device to create more innovative applications, so that technology can truly empower global users!

Click here toHead over to the contest website for more content!