英文:
Is it possible to use MediaProjection API and process every recorded frame during recording?
问题
我想制作一个应用，模拟飞利浦电视上常见的Ambilight效果。为此，我有一个Arduino控制着LED灯，并在串行端口上监听命令以设置LED灯的颜色。
理论上，我希望我的应用能够在后台录制我的Android电视屏幕。这可以通过使用MediaProjection API来实现。但据我所知，你只能在停止录制后获得视频文件。但我希望能够在每隔n毫秒处理每一帧，或者每隔n秒截取一次屏幕截图。
我的问题： 在录制过程中是否可以访问单个帧，或者是否可以使用MediaProjection API在后台每隔n秒钟截取一次屏幕截图。或者是否有其他实现这一目标的方法？
英文:
I want to build an app which simulates ambilight you know from Philips TV's. For this I have an Arduino that controls my LEDs and listens for commands on the serial port to set the color of the LEDs.
In theory I want my app to record the screen of my Android TV in background. This can be achieved with the MediaProjection API. But as far as I know you can only get the video file after stopping the recording. But I want to process every frame or take a screenshot every n milliseconds.
My Question: Can you access a single frame while recording or is it possible to take a screenshot every n seconds in background using the MediaProjection API. Or is there another way achieving this?
答案1
得分: 1
经过进一步调查，我找到了一个解决方案。您可以在MediaProjection上使用ImageReader，并实现其OnImageAvailable回调函数，该回调函数在每次VSync完成时被调用。然后，您可以从阅读器中提取一个位图。
英文:
After further investigation I found a Solution. You can use a ImageReader on the MediaProjection and implement its OnImageAvailable callback this callback gets called every time a VSync is done. Then you are able to extract a Bitmap from the reader.
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论