MeiCam SDK For iOS  3.10.0
Video frame image retriever

Video frame image retriever refers to extract a video frame image at a certain time. The extracted video image can be used as cover for the video or for other purposes. In the SDK, two ways are provided to extract video frame images.

One is use the NvsStreamingContext createVideoFrameRetriever() interface to creates NvsVideoFrameRetriever object and then to call its getFrameAtTime() to extract the frame.

The other is through the NvsStreamingContext grabImageFromTimeline() to extract the frame.

Comparison between the two methods: The first method obtains the image of the original video frame at a certain moment, and the pixel height of the video frame to be extracted can be set as needed. It can also be used across threads without affecting sdk streaming engines. The disadvantage is that users can't extract the stickers, captions, special effects and other materials that have been added on the timeline. If users want to extract a video that includes these resources, they can only compile the video and extract it as the original video. The second method extracts the video frame image of a timestamp on the timeline, including the added stickers, captions, special effects and other materials, which can be extracted as well. The disadvantage is that the engine is stopped and can only be used inside the UI thread.
For the user, the two methods have their own advantages and disadvantages, and can be selected according to the needs.