Vedanuj Goswami

I am a research engineer at Facebook AI Research (FAIR) working on computer vision and natural language processing. I work with Devi Parikh and Marcus Rohrbach. Earlier I worked with Fernando De La Torre at Facebook AR/VR research team on developing efficient neural networks for image and video analysis for Mobile AR and VR like face tracking, recognition and semantic scene understanding.

I contribute to open source deep learning frameworks like Pythia, Pytorch, Glow.

I did my MS in Machine Learning at Georgia Tech, where I was advised by Ashok Goel. Before my MS, I worked briefly with the Camera R&D team at Samsung Research. I completed my bachelors in Computer Science from NITS, India where I was advised by Samir Borgohain.

Email  /  Github  /  Google Scholar  /  LinkedIn

Research

My research is at the intersection of machine learning, computer vision, and natural language processing. In particular, I'm interested in learning algorithms for multi-modal representation learning from both vision and language inputs.

12-in-1: Multi-Task Vision and Language Representation Learning
Jiasen Lu*, Vedanuj Goswami*, Marcus Rohrbach, Devi Parikh, Stefan Lee
* equal contribution
arxiv / bibtex

Unsupervised Image-to-Video Clothing Transfer
Albert Pumarola, Vedanuj Goswami, Francisco Vicente, Fernando De la Torre, Francesc Moreno-Noguer
International Conference on Computer Vision (ICCV) Workshops, 2019
website

Only Time Can Tell: Discovering Temporal Data for Temporal Modeling
Laura Sevilla-Lara, Shengxin Zha, Zhicheng Yan, Vedanuj Goswami, Matt Feiszli, Lorenzo Torresani
arxiv / blog / bibtex

Knowledge Extraction and Annotation for Cross-Domain Textual Case-Based Reasoning in Biologically Inspired Design
Spencer Rugaber, Shruti Bhati, Vedanuj Goswami, Evangelia Spiliopoulou, Sasha Azad, Sridevi Koushik, Rishikesh Kulkarni, Mithun Kumble, Sriya Sarathy, Ashok Goel
International Conference on Case Based Reasoning (ICCBR), 2016
pdf / blog / poster / bibtex

Grammarless Language Generation Algorithm Based on Idiotypic Artificial Immune Networks
Vedanuj Goswami, Samir Borgohain
Australasian Conference on Artificial Life and Computational Intelligence (ACALCI), 2015
code / bibtex

Teaching

GaTech CS7637: Knowledge Based Artificial Intelligence , Fall 2015, 2016, Spring 2016 - Teaching Assistant (TA)


Template from here