Feature Request: "Guided View" Mode (Auto-Panel Detection & Zoom)

[5f20]

Preflight Checklist

• I have searched the issue tracker for a feature request that matches the one I want to file, without success.

Describe your suggested feature

Reading full pages on a desktop monitor often makes text too small to read comfortably. Using manual zoom or the magnifying glass is tedious and breaks the immersion of the story.

I would like a "Guided View" mode exactly like Marvel Unlimited or ComiXology. This feature should:

Auto-Detect Panels: Use image analysis to identify panel boundaries on the page.

Smart Navigation: When pressing the arrow key or spacebar, the camera should zoom in to fill the screen with the next specific panel, rather than just scrolling the page.

Reading Flow: It must support both Left-to-Right (Comics) and Right-to-Left (Manga) ordering.

Focus Mode: Ideally, dim or hide the artwork outside the active panel.

This would allow for a "hands-off" cinematic reading experience where I can sit back and just press one key to advance the story, which is currently the main feature missing compared to premium tablet apps.

I realize this is a complex feature to implement, but it is the "Holy Grail" of digital comic reading. Adding this would elevate OpenComic from a great file viewer to the undisputed best comic platform on any operating system, surpassing even paid commercial options. Please, please consider prioritizing this—it is the one missing piece that would make this software perfect and irreplaceable for the community!

Additional Information

No response

Sample file

No response

[ollm]

Thank you for the suggestion.

I like the idea, but as you mentioned, it's somewhat complicated to implement, mainly the detection of panels (which would probably require using AI to detect them).

At the moment, I've put OpenComic development somewhat on hold due to limited availability and because I'm working on other projects. With the time I do have, I'm focusing on adding AI-related features to OpenComic (artifact removal, descreening, and upscaling) and training the required models, which is taking longer than I initially expected.

I'll try to investigate whether it's very complicated or feasible to detect panels, and I'll try to prioritize this for the next major OpenComic feature.