Video Representation Learning for Fine-Grained Scene Recognition and Retrieval.
Model Learning with limited training data.
Extract structured information from unstructured documents.
Enrichment of educational texts with supplementary information.
Problem Statement: Given a small subset of input samples, collect corresponding output samples from the user and learn a transformation routine that converts provided data into user-intended format.
Motivation: The idea is to facilitate end users, mostly non-experts, with a Programming-by-example, PbE system that captures the user intention and applies to all the provided samples with minimum intervention.
Provide visual aid for a sequence of text based instructions.