Bio

As of Winter 2016, I am a Master's student at IIIT Delhi working with Dr. Chetan Arora (IIIT Delhi) and Dr. M. Balakrishnan (IIT Delhi). I work on Deep Learning and its applications in Computer Vision. I am also interested in Geo-spatial Data Mining.

My Projects

  • Mobility Assistant for Visually Impaired (MAVI)

    • MAVI is an ambitious project in collaboration with Indian Institute of Technology, Delhi (IITD ) aimed at enabling mobility for visually impaired individuals, especially in India. A major challenge in this regard is the wide range of complexities that arise in the Indian scenario due to non-standard practices. Out of its four key Computer Vision components, I am working on animal detection.
    For more information about MAVI, please visit:
    http://www.cse.iitd.ac.in/mavi
    Work done in animal detection is documented here .
    Advisors: Dr. Chetan Arora, Dr. M. Balakrishnan


  • Elephant Accident Prevention

    • This project is an effort towards saving elephants from getting hit by high-speed trains passing through national park corridors. We are using time-series data analysis to classify train and elephant seismic data.
    Advisor: Dr. P.B. Sujit


  • Autonomous E-Rickshaw

    • This is an endeavour to achieve last-mile connectivity in India. I am working on its Computer Vision component that involves lane detection and obstacle detection using LiDAR.
    Advisor: Dr. P.B. Sujit


  • Ego-centric Action Recognition using Sparse Flow and LSTM

    • We are using the HUJI dataset to perform action recognition. The approach involves determining sparse flow between the frames of a video and feeding it to an LSTM network.

  • Recommendation System for Regional Indian Cinema

    • There's no recommendation system in place for Indian regional cinema. I have designed an online portal to collect user ratings for Indian movies and their demographics. This data, once collected, will be used to build a recommender.

  • Multimodal Search Engine

    • This search engine alows the user to express a search query through multiple modes (text, image and audio). It uses deep learning to understand images and text queries.

Publications


Discovering Spatial Regions of High Correlation

Prerna Agarwal*, Richa Verma*, Venkata M. V. Gunturi
ICDM 2016 Workshop

Ontology Based Disease Diagnosis System with Probabilistic Inference

Prerna Agarwal*, Richa Verma*, Anupama Mallik
IICIP 2016

Contact Me

richa15054@iiitd.ac.in
https://github.com/richa-verma