Handwritten-Word-Spotting-with-Corrected-Attributes

Computer Vision (Spring2020 CSE578) Semester Project - Handwritten Word Spotting with Corrected Attributes

An implementation of Almazan's 2013 ICCV paper

Abstract

The project’s focus is to provide an approach to multi-writer word spotting, where the goal would be to find a query word in a dataset comprised of document images. It is an attributes-based approach that leads to a low-dimensional, fixed-length representation of the word images that is fast to compute and, especially, fast to compare. This approach would lead to a unified representation of word images and strings, which seamlessly allow one to indistinctly perform query-by-example, where the query is an image, and query-by- string where the query is a string.

Team Members

1. Ritvik Agrawal
2. Vivek Chandela
3. Kirandevraj R