What I do is only look at a small part of every picture and when recalling I know which part of the given picture is familiar and based on that I can put it into the proper place.
I use a 4x4 grid and the first 16 picture parts are memorized as squares off this grid. The next ones as 4 rows, 4 columns and 4 quadrants and the last 2 pictures go into the short term memory.
With this method I was able to get under 1 minute for 30 pictures while with the 2/loci method I needed it way more time.
Is it a known method?