價格：免費
更新日期：2019-04-23
檔案大小：253.5 MB
目前版本：1.1
版本需求：需要 iOS 11.2 或以上版本。與 iPhone、iPad 及 iPod touch 相容。
支援語言：英語
Since iOS 11 Core Machine Learning enables every iOS User to use artificial intelligence machine learning.
- Detect objects and scenes
- Choose machine learning models
- Detect faces and highlight the landmarks in a face
- Switch between front and back camera
- Voice output
- Online connection not needed - everything runs locally
General:
After the first start you are in object detection mode. Point your camera to objects and the detection starts automatically. In the top right you can tap on the settings icon.
For face detection you don't have to choose a model - just tap on 'Face detection' to switch to face detection mode and on 'Object detection' to return to object detection.
Settings:
In the settings you can choose a machine learning model from the list. The first four models are for object detection. The model 'Places205' is for scene detection from 205 categories such as airport, bedroom, forest, coast and more.
Tap done when you have selected a model. You return to the main screen.
Speech can be turned off completely.
Use the 'Send feedback' button to write me an e-mail with your impressions or suggestions.
Tip: The detection works bests in lighted environments. You can also search photos of things or face in the internet and just point your iOS camera to the photos.
Object detection:
Point your camera to objects. The app will try to detect the object from more than 1000 categories. Try animals, furnishing, household objects, cars, fruits and vegetables. The 1st guess is the most likely object description with the highest confidence. The percentage value of confidence is stated right to the object description.
The 2nd most likely guess is also stated.
With the slider "Speech threshold" you can set an percentage value. When the confidence percentage is greater or equal as that value the object will be spoken. When you set the slider all way to the left speech is disabled.
In the top banner in the camera feed the current machine learning model is on the left. The frames per second of detection on the right of the banner.
Face detection:
Point the camera at a face or a group of people. For every face the major landmarks like lower face contour, nose, eyebrows and eyes are marked in yellow. The lines moves as to move the iOS device.
Face detection only works in portrait mode. You can switch to the front camera with the camera icon in the upper right.
支援平台：iPhone, iPad