Obviously, the bigger the image, the easier to detect faces.
The API uses eyes as the "facial detection" and there is no other detail (face dimensions mouth location etc.).
Also faces that are taken from the side will not be detected. Sunglasses/glasses turns out to interfere with detection.
It seems to take the same amount of time regardless of how many faces you specify in the FaceDetector constructor.
I recommend the picture size smaller than 800x800 pixels coz it will take a lot of time to process the large image.
The following app runs the image processing code in a separate thread, and notifies the UI thread when the processing gets done, then calls invalidate() to redraw the canvas.
(画像解析のコードは別スレッドで走らせて、処理が終了したらUIスレッドに知らせてキャンバスを描き直しています。)
特に何でもないアプリですが、
もしコード見たい方がいたらアップしますので、コメントでお知らせください。
Demo Video:
The API uses eyes as the "facial detection" and there is no other detail (face dimensions mouth location etc.).
Also faces that are taken from the side will not be detected. Sunglasses/glasses turns out to interfere with detection.
It seems to take the same amount of time regardless of how many faces you specify in the FaceDetector constructor.
I recommend the picture size smaller than 800x800 pixels coz it will take a lot of time to process the large image.
The following app runs the image processing code in a separate thread, and notifies the UI thread when the processing gets done, then calls invalidate() to redraw the canvas.
(画像解析のコードは別スレッドで走らせて、処理が終了したらUIスレッドに知らせてキャンバスを描き直しています。)
特に何でもないアプリですが、
もしコード見たい方がいたらアップしますので、コメントでお知らせください。
Demo Video:
Leave a comment