I have a series of images, each containing a single word from a known dictionary of 2048 words. The size, font, and position of the word is known ahead of time, and I simply need to tell which word from the dictionary is in the image.
Standard OCR tools like tesseract are proving surprisingly bad at this task, and I'm wondering if it's practical to build my own solution from a RNN library or something similar.
Are there good resources for getting started with this sort of task for someone with absolutely no prior machine learning knowledge or experience?
