Pull Request Readiness Checklist

See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request

• I agree to contribute to the project under Apache 2 License.
• To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
• The PR is proposed to the proper branch
• There is a reference to the original bug report and related work
• There is accuracy test, performance test and test data in opencv_extra repository, if applicable
  Patch to opencv_extra has the same branch name.
• The feature is well documented and sample code can be built with the project CMake

Motiviation

Let's image use case where execution should last: 50ms

0ms       10ms      20ms      30ms        40ms       50ms
 |---------|---------|----------|----------|----------|

and one pipeline iterations takes 15ms.

In that case the picture will be:

       0 frame          1 frame       2 frame         3 frame
 |----------------|---------------|---------------|---------------|

0ms       10ms      20ms      30ms        40ms         50ms       60ms
 |**********|**********|**********|**********|**********|**********|
 
 FPS = 4 frames * 1000 / 60ms = 66.6 FPS 

But if user enabled --drop_frames the picture would be different:

       0 frame               1 frame              2 frame
 |----------------|    |---------------|      |---------------|

0ms       10ms        20ms      30ms        40ms         50ms  55ms
 |**********|*****-----|**********|*****------|**********|*****|
 
 FPS = 3 frames * 1000 / 55ms = 54.54 FPS 

So --drop_frames could significantly affect FPS, because number of iterations depends on this flag.
But what if user want to run certain number of iteration but at the same time schedule pipelines with --drop_frames mode?

Implementation

Now every Pipeline could have local num_iters property. The logic is following:

• num_iters is local property for Pipeline. (no global num_iters)
• if num_iters is set then it takes priority over work_time.
• work_time is still global (no local work_time per Pipeline), but now is not necessary to be specified.
• If Pipeline doesn't have num_iters property and global work_time isn't specified then throw error.