The original UI prototype for this extension included the concept of a "feed" of images (~25 at a time, with the ability to load more as desired by clicking a button).
The latest designs from @PDrouin-WMF have moved away from this. Instead, the metaphor is that of a stack of cards – the user sees one at a time, has the ability to skip to the next if desired, and sees some visual elements that imply additional cards are stacked beneath the one they are presently looking at. Individual cards should be responsive and the entire UI should feel very comfortable on mobile devices as well as on desktop.