A Taste of TensorFlow on My Android Phone (II)

This post is a follow-up to my last post: A Taste of TensorFlow on My Android Phone.

You could download my compiled Apk file here and install on your android device (>Android 7.0) (22-May-2018).

Update: 25-May-2018, fixed the bug of object tracking function in TF Detect. PS: While running the activities, pressing the volume keys on your device will
toggle debug visualizations on/off, rendering additional info to the screen that
may be useful for development purposes.

 

Feel free to let me know if there are any bugs 🙂

TensorFlow Mobile

In the last post, I just tried the TensorFlow for object classification (TF Classify). This time I installed all four demos of the TensorFlow Mobile for Android according to this tutorial: TensorFlow Lite Demo for Android. They are awesome 😛

 

Let see two photos of my desk, the TF Detect App find out my keyboard, laptop, mouse, coffee cup and my “TVs” :).

Screenshot_2018-03-06-09-45-41-633_org.tensorflow.demoScreenshot_2018-03-06-11-20-48-577_org.tensorflow.demo

Here is another photo about detecting cars on the street. I know It is not perfect now but it is a good start!

Screenshot_2018-03-06-07-30-21-470_org.tensorflow.demo

Here are the examples of TF Classify, TF Stylize, and TF Speech.

Screenshot_2018-03-06-09-44-57-946_org.tensorflow.demoScreenshot_2018-03-06-09-44-32-468_org.tensorflow.demoScreenshot_2018-03-06-09-44-42-265_org.tensorflow.demo

Here are more screen captures from my XIAOMI 4c Phone:

 

 

 

This slideshow requires JavaScript.

This is the introduction from the official documentation:

The TensorFlow Lite demo is a camera app that continuously classifies whatever it sees from your device’s back camera, using a quantized MobileNet model.
You’ll need an Android device running Android 5.0 or higher to run the demo.
To get you started working with TensorFlow Lite on Android, we’ll walk you through building and deploying our TensorFlow demo app in Android Studio.
#1. TF Classify uses the Inception v3 model to label the objects it’s pointed at with classes from Imagenet.
#2. TF Detect uses a multi-box model to try to draw bounding boxes around the locations of people in the camera.
#3. TF Stylize implements a real-time style transfer algorithm on the camera feed. You can select which styles to use and mix between them using the palette at the bottom of the screen, and also switch out the resolution of the processing to go higher or lower rez.
#4. TF Speech recognizes few words from the mic.

I will upload the apk files later ~ Cheers!

-END-

Author: Caihao (Chris) Cui

Data Scientist and Machine Learning Expert: Translating modern machine learning and computer vision techniques into engineering and bringing ideas to life to design a better future.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s