Machine simulation of human reading has caught the attention of computer science researchers since the introduction of digital computers. Character recognition is the process of recognizing either printed or handwritten text from document images and converting it into machine-readable form. Character recognition is successfully implemented for various foreign language scripts like English, Chinese and Latin. In the case of Indian language scripts, the character recognition process is comparatively difficult due to the complex nature of scripts. MODI script - an ancient Indian script, is the shorthand form for the Devanagari script in which Marathi was written. Though at present, the script is not used officially, it has historical importance. MODI character recognition is a very complex task due to its variations in the writing style of individuals, shape similarity of characters and the absence of word stopping symbol in documents. The advances in various machine learning techniques have greatly contributed to the success of various character recognition processes. The proposed work provides an overview of various feature extraction and classification techniques used in the recognition of MODI script till date and also provides evaluation and comparison of these techniques.