Flutter Landmark Classifier (TFLite) - SpeedCode

preview_player
Показать описание
😀Thanks for watching :D
This is a project that can recognize and search landmarks in each region.
(Asia, South America, North America, Europe, Oceania, Antarctica)

It includes how to apply the model and how to process label data.

Unfortunately, the performance of the tflite model results worse than the example of tfhub.
To be used in real-world projects, you need to improve the performance of your model.

If you are in an iOS environment, you must add the following in Podfile:

▶ Label Data Edit Page

▶ Model Info
* TFHub Landmark Models

▶SourceCode
* Github

▶Packages
* multi_image_picker

* image

* tflite_flutter

* tflite_flutter_helper

* url_launcher

▶ Keyword
Classification, Landmark, Flutter, Tensorflow, TFLite

▶Timestamp
00:00 - Intro
00:44 - Project Setting
01:01 - AndroidManifest Setting
01:08 - Podfile Setting (TFLite Setting)
01:30 - TFLite Model & Label Setting
02:12 - Home Page
10:28 - ImageService
11:30 - ClassificationService
18:45 - Classification Page

#flutter #tflite #classification
Рекомендации по теме
Комментарии
Автор

Does it work in release mode? Because I had trouble running apps with tflite in release mode?

andrejvujic
Автор

Cloned the project to my laptop, tried running it on a Samsung android device, it says permission denied when I click on the camera icon for the image picker to pick image from gallery. Can you help me out?

nakshatrasingh
welcome to shbcf.ru