I've [uploaded a picture][1] to Wolfram's ImageIdentify of graffiti on the wall, but it recognized it as 'monocle'. Secondary guesses were 'primate', 'hominid', and 'person', so not even close to 'graffiti' or 'painting'.

Is it by design, or there are some techniques to teach a convolutional neural network (CNN) to reason and be aware of graffiti based on the bigger picture context? Currently it seems as if it's detecting literally *what* it is, rather than *why* it's there.




[![Wolfram's Image Identify: monocle/graffiti][2]][2]


  [1]: https://www.imageidentify.com/result/0lkzuttdxipub
  [2]: https://i.sstatic.net/akquMm.png