Statistics are calculated for those time steps where a mask is provided, for the others not. When the view asks for the statistics an exception is thrown if the statistics don't exist.

see: QmitkImageStatisticsCalculator, line 209ff (Histogram visualization)

We put a try catch around that . There seems to be another exception thrown when willing the table view. The statistics table is not filled:

Pushed new branch T26029-4DImageAnd4DSegmentation.

Solution: GetStatisticsForTimeStep method now returns an empty StatisticsObject if no mask is provided for this time step. Thus, histogram and table are empty for those time steps.
Discussion: 1. 4D image + 3D segmentation

2. 4D image (5 time steps) + 4D segmentation (1 time step)

@hentsch Is the one time step segmentation applied to all time steps? What is the expected behavior?

I have concerns with the latest commit. See comments there. You can come by and ask if you have any questions.

@hentsch Is the one time step segmentation applied to all time steps? What is the expected behavior?

Yes. One segmentation timestep to rule all image timesteps! But the last time I tested it, it worked fine.

@kausch @kleina you resolved this task, right?
If not done already, please merge to the master.

Yeah, we did... to be honest, I don't know if it is merged yet. @kausch can you please check this?

I merged it, we only pushed the branch

In T26029#181806, @kausch wrote:

I merged it, we only pushed the branch

@kausch You have not closed the task ;). Was this task done in its own branch or as a "side" task in an other branch?