Photo straightening

Hi all,

Sports photographer here, working exclusively in JPEG due to very tight turnaround times. A typical event easily generates 10k+ images, so the entire pipeline has to be highly optimized and as close to zero-touch as possible.

I’m currently evaluating Retrobatch for batch processing, specifically in a fully automated workflow context.

One key bottleneck for me is straightening. In fast-paced environments, a percentage of frames inevitably come out slightly tilted, and manual correction at this scale is not viable.

From a technical standpoint: does Retrobatch offer any form of reliable auto-straighten based on horizon detection or dominant line analysis that can be applied consistently in batch, without per-image intervention?

Ideally, I’m looking for something deterministic enough to trust across large datasets, not just occasional heuristic corrections.

If anyone is running Retrobatch in high-volume production (sports/events) and has tested this, I’d appreciate insights on accuracy, edge cases, and whether it’s robust enough for unattended workflows.

Thanks,

Filipe

Retrobatch has an “Auto Level” node in the Transform category. Have you tried that out?

And if it’s not working well for you, I’d love to take a look at some images that it’s having troubles with (you can send them to support@flyingmeat.com).

-gus

Hi Gus, thanks for your fast reply. I haven’t bought Retrobatch yet hence the questions.

I’ll download and try it this weekend and will let you know.

Thanks!

Filipe