Reply to thread

Message: <blockquote data-quote="ziphnor" data-source="post: 164328" data-attributes="member: 14215">The autocropper only computes the bounds of the actual video in the frame (no cropping of actual video content!), it doesnt decide how to display it, i leave that to the Zoom modes defined in Geometry.cs.So, if you mean a 21:9 letterbox in a 4:3 transmission (without subs), then it wouldnt get cropped left and right (depending on the active zoom mode). The autocropper would compute a 21:9 bounding box and if you use the Normal zoom mode then you would still get some black bars top and bottom (because its wider than the 16:9 display). If you use the Stretch mode you would get a fullscreen image with people looking a bit tall, while Zoom would crop the 21:9 image to a 16:9 image by cutting left and right sides off.Thats also why it could detect 21:9 or 16:9 material (without subs), because the resulting bounding box will have those approx. aspect ratios.Earlier on i wrote the autocropper in C++ and actually had some experimental code there for autodetecting subtitles and moving them into the letterbox, allowing for better zoom. It was costly though, and was also a bit jittery <img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7" class="smilie smilie--sprite smilie--sprite2" alt=";)" title="Wink    ;)" loading="lazy" data-shortname=";)" />Basically i was thinking that if we were to merge, i would simply take care of the frame analysis and you use the result in which ever way you see fit <img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7" class="smilie smilie--sprite smilie--sprite1" alt=":)" title="Smile    :)" loading="lazy" data-shortname=":)" />Btw, as you mention we might want to discuss this elsewhere and you should also be aware that i will be travelling (combined business and holiday) starting from a few days and lasting about a month, so things will probably have to wait until i get back.</blockquote>

Verification