Short update to automated assessments

One question I got about the results I presented at the end of my last blog entry was that while the results are interesting, it would be more interesting to perform the comparison on a note level. Very true! So here we go:

The example is again TP12, and here are the detailed results on a symbol-by-symbol (notes and rests) level:

OMR3 had looked very good on a summarized level in the last comparison. Here is the detailed comparison:

Nice! While the first comparison for OMR3 showed 3 errors, it turned out that the ground truth contained two errors. OMR2 is a different story, since it added a measure, which of course makes a symbol-by-symbol comparison difficult. (Technically, I compare on a measure level.) More about that next week!

One Comment Add yours

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s