Anubhab Sen

I am an undergraduate student at IIIT, Hyderabad. At IIIT, I work at the Centre for Visual Information Technology with Prof. Avinash Sharma, where I have worked on optical flow, 3D reconstruction of non rigid objects, sketch based image retrieval and gaze following.
I am also working with Prof. Ponnurangam Kumaraguru to analyse and try to predict trends on Twitter regarding the Indian general election, 2019.
I am working with Prof. Vinoo Alluri to investigate connection between the musicality and various cognitive abilities and personality traits.

During the summer of 2018, I worked at the Computer Vision Lab, at Indiana University Bloomington under the guidance of Prof. David Crandall on a system that can classify notes played and finger used by a piano player using the video and audio feeds. Now I am working on preserving privacy of non-subjects in photographs using GANs.

I've spent time at Gibbr working on a chat bot, Felicity Web Team (IIIT Hyderabad) working on the portals for online contests for the college annual fest, IGCAR, DAE, India in the computer division.

Email  /  CV  /  GitHub  /  LinkedIn

Sections
Research

I'm interested in computer vision, machine learning (deep learning in particular) and image processing. I have been working on 3D reconstruction of non rigid objects, gaze following, sketch based image retrieval and depth estimation using unsupervised methods.
Some papers I am currently working with:

Unsupervised Learning of Depth and Ego-Motion from Video
Tinghui Zhou, Matthew Brown, Noah Snavely, David Lowe
CVPR, 2017
project page

Unsupervised learning framework for monocular depth and camera motion estimation from unstructured video sequences

3D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstruction
Christopher B. Choy, Danfei Xu, JunYoung Gwak, Kevin Chen, Silvio Savarese
ECCV, 2016
project page

Generate voxelised reconstruction of object given one or multiple views

Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
Jiajun Wu, Chengkai Zhang, Tianfan Xue, William T. Freeman, Joshua B. Tenenbaum
NIPS, 2016
project page

Generates 3D objects from a probabilistic space by leveraging recent advances in volumetric convolutional networks and generative adversarial nets.

Where are they looking?
AdriĆ  Recasens, Aditya Khosla, Carl Vondrick, Antonio Torralba
NIPS, 2015
project page

Given an image and the location of a head, the gaze of the person is followed and the object being looked at is identified.

Course Projects
sbir

Sketch based image retrieval
Anubhab Sen, Rohit Gajawada, Aditya Bharti, 2017


Given a sketch, search a database from images which are similar to the sketch and display them.

ultimate-tic-tac-toe

Ultimate tic tac toe bot (4x4x4x4)
Anubhab Sen, Rohit Gajawada, 2017


An AI bot to play a variation of ultimate tic tac toe (4x4x4x4) using minimax and alphabeta pruning along with iterative deepening and a heuristic to handle time limits.

aquarium

WebGL Aquarium
Anubhab Sen, Megh Parikh, Sourabh Rajguru, Arun Garimella, 2016


An aquarium made using completely using WebGL and free resources like models and textures.

cshell

C-Shell
Anubhab Sen, 2017


A bash clone implemented in C. Some bash commands are implemented and the rest are invoked using execvp. A few additional commands such as pinfo, jobs, kjob and kill have been implemented to enhance productivity.

Teaching
graphics

Machine Learning (Monsoon 2018) - Teaching Assistant

Course with over 250 students including Bachelors, Masters, PhD and professionals registered.




graphics

Computer Graphics (Spring 2018) - Teaching Assistant

Course with over 150 undergraduates registered.




os

Operating Systems (Monsoon 2017) - Teaching Assistant

Course with over 150 undergraduates registered.





Source: Jon Barron