Viraj Prabhu

I am a second year M.S. student majoring in Computer Science (specializing in Machine Learning) at Georgia Tech. I am advised by Prof. Devi Parikh and also work closely with Prof. Dhruv Batra.

My broad research interests lie at the intersection of computer vision and natural language processing, particularly, in developing interactive multimodal AI agents that can be deployed in the real world. In the past, this has included research on equipping visual conversational agents with mechanisms for question relevance detection, studying the utility of various interpretability modalities proposed for such agents in the context of human-AI teams, and exploring human-in-the-loop evaluations of such agents.

I spent the last summer as a research intern at Curai, a young machine learning-healthcare startup, where I worked on automated dermatological diagnosis in a few-shot learning setup and was mentored by Anitha Kannan. Prior to joining Georgia Tech, I spent a wonderful year as a research assistant in the Machine Learning and Perception lab at Virginia Tech, advised by Prof. Dhruv Batra. Previously, I worked as a software developer at Adobe Systems India on Adobe Captivate Prime, a Learning Management System for enterprise.

I received my Bachelor's degree in Computer Science from BITS Pilani. Over the course of my undergraduate studies, I was fortunate to undertake research internships at Adobe, CEERI Pilani, and Tonbo Imaging, where I worked on problems ranging from video segmentation to automated camera calibration. For a detailed list, please take a look at my CV.

On the side, I am an open-source contributor to CloudCV and served as a mentor for Google Summer of Code (2016, 2017) and Google Code-In (2017) for the Fabrik project. I also enjoy running, soccer, playing the guitar, and (occasionally) writing.

Feel free to reach out to me at



  • Among the top 30 percent highest scoring reviewers for NeurIPS 2018.

  • Winner of VT Hacks 2017, a Major League Hacking event.

  • Awarded travel scholarship to represent CloudCV at Google Summer of Code Mentor Summit 2016, 2017.

  • Winner, Google Hackathon, APOGEE 2014 (BITS Pilani's annual technical symposium).

  • Second Place, Project Presentation (Adaptive Technology and Design Appliances Tracks), APOGEE 2013.

  • Top 0.2%, BITSAT 2011 (from >120k students).

  • Top 0.1%, ICSE 2009 (from >150k students), awarded Amul Vidya Shree.


Prototypical Clustering Networks for Dermatological Disease Diagnosis


Viraj Prabhu, Anitha Kannan, Murali Ravuri, Manish Chablani, Xavier Amatriain


Fabrik: An Online Collaborative Neural Network Editor


Utsav Garg, Viraj Prabhu, Deshraj Yadav, Ram Ramrakhya, Harsh Agrawal, Dhruv Batra

[Paper] [Code]

Do Explanations make VQA Models more Predictable to a Human?

(EMNLP 2018)

Arjun Chandrasekaran*, Viraj Prabhu*, Deshraj Yadav*, Prithvijit Chattopadhyay*, Devi Parikh
(* equal contribution)


PyTorch implementation of Learning Cooperative Visual Dialog Agents with Deep Reinforcement Learning

Nirbhay Modhe, Viraj Prabhu, Michael Cogswell, Satwik Kottur, Abhishek Das, Stefan Lee, Devi Parikh, Dhruv Batra


The Promise of Premise: Harnessing Question Premises in Visual Question Answering

(EMNLP 2017)

Aroma Mahendru*, Viraj Prabhu*, Akrit Mohapatra*, Dhruv Batra, Stefan Lee
(* equal contribution)

[Paper] [Code] [Dataset]

Evaluating Visual Conversational Agents via Cooperative Human-AI Games

(HCOMP 2017)

Prithvijit Chattopadhyay*, Deshraj Yadav*, Viraj Prabhu, Arjun Chandrasekaran, Abhishek Das, Stefan Lee, Dhruv Batra, Devi Parikh

[Paper] [Code]

It Takes Two to Tango: Towards Theory of AI's Mind

(Chalearn Looking at People Workshop, CVPR 2017)

Arjun Chandrasekaran*, Deshraj Yadav*, Prithvijit Chattopadhyay*, Viraj Prabhu*, Devi Parikh
(* equal contribution)

[Paper] [Demo] [Code]


Graduate Research Assistant, Visual Intelligence Lab, Georgia Tech, advised by Prof. Devi Parikh (Fall 2017 - Present)

Working on problems at the intersection of computer vision and natural language processing.

Research Intern, Curai, mentored by Anitha Kannan (Summer '18)

Worked on few-shot learning for automated dermatological diagnosis, proposing approaches to to model intra-class diversity and long-tails present in dermatology datasets.

Research Scholar, Machine Learning and Perception Lab, Virginia Tech, advised by Prof. Dhruv Batra (Fall 2016 - Spring 2017)

Worked on deep learning approaches to Visual Question Answering. Also served as teaching assistant for Intro to Machine Learning in Fall 2016, taught by Dr. Stefan Lee.

Mentor, CloudCV, Google Summer of Code 2016, 2017, and Google Code-In 2017

Mentored and presently help maintain the Google Summer of Code project Fabrik, a web platform to collaboratively design, edit, and share deep neural networks from within a browser. [GitHub]

Member of Technical Staff, Adobe Systems India (Jul 2015 - August 2016)

Individually responsible for the Captivate Prime Android app through two release cycles, contributing with features and bugfixes for offline content play-back, syncing and UI.

R&D Intern, Tonbo Imaging (Spring 2015)

Computer vision startup developing sensor systems for battlefields and reconaissance.

Designed algorithm for automated camera calibration and implemented a boresighting algorithm for the company's video precision boresight tool.

Research Intern, Adobe Presenter Video Express (PVX) (Summer 2014)

Developed fast graphcut-based segmentation algorithm for real-time background substitution in video. Transferred into Magic Green Screen, the marquee feature of PVX 11.[Demo] [Blog]

Project Assistant, advised by Dr. Jagdish Raheja (Spring 2013)

Prototyped a modern teleconferencing application that captured a 360° FOV by interfacing multiple Kinect sensors, and identified and displayed the current speaker.


BE Computer Science


MS Computer Science