Hi,

I'm trying to use the Image Sequence Probe for the first time, but I can't seem to get it working properly.

I've read the manual, but I'm still missing something. I must be doing something wrong here.

Here is what I'm doing:
· I loaded a custom CSV into the Characterisation window.
· When I use the Virtual Probe, all patches appear correctly in the graphs after loading.
· However, when I switch to the Image Sequence Probe and load my image sequence (patch%06d.dpx), no points appear in the graphs during or after the load process.
· The image sequence status turns green and indicates that it's loading correctly.
· Offset is set correctly (sequence starts at 000001).
· Step is set to 1.

But no measurement points are generated — nothing appears in the graphs.

It feels like the sequence is recognised, but no actual measurements are being triggered.

Am I missing a required step to initiate the measurement process when using Image Sequence Probe?

Any help would be greatly appreciated.

Thanks!

What are the images labelled as exactly, and what is the text string you are using in full?
And what is the .csv?
And make sure you have Patch Sequence Sorting disabled, so the .csv list order is used.
Also enable Tangents so you can see if the measurements are just not where you expect them to be.

If you still have issues, host the image patches and the .csv somewhere so I can download to test.

Steve

Thank you, Steve! Disabling the path sequence sorting fixed it. All patches are now read correctly with the Image Sequence Probe as well as my regular probe.

I was able to successfully profile and calibrate my displays (profiling + LUT generation), which works great.

However, I'm struggling with generating a LUT for camera matching / film.

For example, as a simple test case to understand the workflow, I used 24 patches of a ColorChecker shot on both a Sony camera and a film camera, with the intention of matching one to the other. I believe I was able to profile this using a custom .csv and the Image Sequence Probe, but I can't seem to generate the LUT correctly.

I've watched a ton of videos and read the relevant section in the guide, but I still can't seem to figure it out. I come from using MatchLight, which performs the kind of transform I'm trying to achieve now with ColourSpace and the Image Sequence Probe

We will need the two patch sets and .csv list you have to perform any assessment.
Please provide a link to them.
Use direct email, as that is easier.

Steve

If you have only used the limited colours on a ColorChecker chart I don't think that would be enough for a good LUT generation.
MatchLight uses a very different approach to the LUT generation so probably would work but would be less accurate.

ColorCal
That is likely correct.
We'll know more when the images and .csv are provided.

Steve