mintograde
New Member
To preface, I don't have any experience working with OBS in a development sense, or with OCR in general, and I'm using this small project as a way to learn both. In other words, this is more of an "is this even possible" kind of post.
My goal is to generate a timestamped log of post-game stats (example here), so I can more easily keep track of when I play games and with whom. The game itself is on a console, and the official stats website does not contain any of the stats I'm looking for, so I believe my only option is to process the video through OBS while streaming/recording.
I'm less concerned about the actual OCR part, as I see that as the "fun" part to learn. My first concern is just getting the image data from OBS so I can start playing with/manipulating it.
I'm not sure where to start, though, so here are some initial questions:
My goal is to generate a timestamped log of post-game stats (example here), so I can more easily keep track of when I play games and with whom. The game itself is on a console, and the official stats website does not contain any of the stats I'm looking for, so I believe my only option is to process the video through OBS while streaming/recording.
I'm less concerned about the actual OCR part, as I see that as the "fun" part to learn. My first concern is just getting the image data from OBS so I can start playing with/manipulating it.
I'm not sure where to start, though, so here are some initial questions:
- Is the OBS stream output exposed to external applications through some API already?
- If not, should I start by making a plugin that, say, takes a snapshot N times per second and either dumps out an image file or allows communication with external applications?
- Should the OCR functionality even be developed as an application external to OBS, or would it be easier to do the OCR stuff as a plugin within OBS?
Last edited: