This project extends some of the material we presented in Project 2, by demonstrating how a simple capture implementation can be extended to provide interesting features. By the end of a code review of this project, you should understand how to use the AVFoundation API to temporarily halt your video capture module, adjust capture settings to use photo-quality resolution, capture a picture, then resume video capture.
Important: To use this application, follow the instructions in the Quick Start section of the main README file for this repository.
Since we are importing a number of classes implemented in project 2, the header search paths in the project build settings must be extended to look in the project 2 directory. You will notice an extra group in this project's navigator space with references to the files we need.
-
The only new implementation in this project is the TBExamplePhotoVideoCapture class. By subclassing the video capture module implemented in project 2, we save some time setting up standard video capture and focus only on manipulating AVFoundation to give us a picture in the middle of a (video) capture session.
-
In testing, we noticed that the image sensor takes a moment to adjust white balance and exposure after switching to the capture session preset for photo quality (see
pauseVideoCaptureForPhotoin TBExamplePhotoVideoCapture). The example implementation stalls the photo capture with a busy loop, but a more sophisticated approach might be considered to ensure the best experience for the end user. -
This implementation briefly pauses the video feed during image capture. Note that a prolonged delay to the video capture consumer might result in other subscribers timing out the stream. If you wait for too long to send consecutive video frames, you might lose the publisher altogether! Consider sending a freeze frame or even a blank image buffer to the video capture endpoint if you need to pause video for a long (>2 second) period.
-
An alternative approach to this problem might be to continuously pipe photo- quality video into the video capture consumer. This might not work on some devices, based on processing capability and the fidelity of the image sensor. Feel free to experiment and let us know how it goes for you!
-
Note that this sample application is not supported in the XCode iOS Simulator because the custom video capturer needs to acquire video from an iOS device camera.