Hacker News Reader

Comment by andrewstuart | Hacker News Reader

You could do this with ffmpeg on a screenshot PNG.

ffmpeg -i screenshot.png -vf \ "crop=iw:1080:0:n*(20000-1080)/600,format=yuv420p" \ -t 10 -r 60 output.mp4

Title of the post fails to describe what this fully does and you didn't seem to have read the repo. This generates a live interaction with the website, with scrolls, timings, clicks, route navigations, etc. A mixture of AI generated cues that you can use via skills + scripts to be feed with those cues that will process the video automatically via headless chromium and ffmpeg. Seems really interesting

What problem does this solve?

Many websites have looping or scroll-triggered animations, or play gifs / videos that a static image wouldn't capture.

i bet you its in aid of advertising somehow

Say you have documentation with a video showing the user how to open a page, click a button, scroll, etc. Instead of having to re-record those videos every time you update the UI you can use this library to automatically recapture it on every push.

Seems pretty useful to be able to write your documentation videos as steps and generate new videos everytime the platform changes. Let’s say you have 200 videos showing how to do different things and now the UI changes.

You can either leave the videos as is (and get more confused customers) or you can spend a lot of hours recording 200 videos again. Or you can run this and get the 200 videos done in the background. Let’s say that your app is changing every month because you are early in your iteration. That is a lot of time saved every month.

Or you want to produce demo videos in multiple languages.

That’s a great usecase too.

Sent you an email.

Reply on news.ycombinator.com