Shankhanil Ghosh

Me, in front of the TUIT, Tashkent, 2022

Hi, I am Shankhanil Ghosh, a computer vision engineer at BigThinx, and I build novel deep learning pipelines for the fashion tech industry. My research interests, apart from Computer Vision also lie in multimodal deep learning, Natural language process and generative AI. I have 3+ years of experience working in deep learning, including both academia and industry.

I have a master’s degree in Information Technology from University of Hyderabad, and a bachelor’s degree in Computer Science and Engineering from University of Calcutta. I had started my research journey working with social Media NLP , where I developed solutions to use tweets to predict traffic scenarios at a location. The primary focus of the idea was to use traffic-intelligence from tweets to build traffic management heuristics. I am deeply invested in multi-modal deep learning because of the because of the sheer possibilities it has. I have worked on building affective computing solutions using deep learning. The foundation of this work began in June, 2021, when I was working on building ReSenseNet, a multimodal deep learning architecture for sentiment analysis. Currently, I am working on 3D reconstructure methodologies for the fashion-tech industry.

I have also invested myself in building startups. In the height of COVID in 2020, small F&B business owners and their customers faced a dilemma regarding food and catering services. I, along with a few friends, launched Connet-NoTouch, a web application product that helped such businesses function in a no-contact manner.

news

Nov 17, 2022	Delivered a technical turotial at Tashkent University of Information Technologies at Tashkent, Uzbekistan on “Introducing deep learning models for human emotion recognition and Analysis” at The 14th International Conference on Intelligent Human Computer Interaction (IHCI-2022), on 21st October, 2022
Aug 8, 2022	Joined BigThinx as a Computer Vision Engineer
Dec 22, 2021	(Virtually) Delivered a technical turotial on “What is that facial expression? Exploring human facial and pose from video” as a part of “Understanding emotion for depression and anxiety detection from text, audio and video using machine learning : a hands-on tutorial”, The 13th International Conference on Intelligent Human Computer Interaction (IHCI-2021), 22nd December, 2020
Nov 24, 2021	Our paper “reSenseNet: Ensemble Early Fusion Deep Learning Architecture for Multimodal Sentiment” was accepted to the 13th International Conference on Intelligent Human-Computer Interaction (IHCI-2021)
Sep 17, 2021	(Virtually) Presented our paper “Speech@SCIS:Annotated Indian video dataset for speech-face cross modal research” to the International Conference on Smart Computing and Informatics (SCI-2021)

selected publications

reSenseNet: Ensemble Early Fusion Deep Learning Architecture for Multimodal Sentiment Analysis

Ghosh, Shankhanil, Saha, Chhanda, Molakathala, Nagamani, Ghosh, Souvik, and Singh, Dhananjay

In Intelligent Human Computer Interaction: 13th International Conference, IHCI 2021, Kent, OH, USA, December 20–22, 2021, Revised Selected Papers 2021

Bib
@inproceedings{ghosh2022resensenet, bibtex_show = {true}, title = {reSenseNet: Ensemble Early Fusion Deep Learning Architecture for Multimodal Sentiment Analysis}, author = {Ghosh, Shankhanil and Saha, Chhanda and Molakathala, Nagamani and Ghosh, Souvik and Singh, Dhananjay}, booktitle = {Intelligent Human Computer Interaction: 13th International Conference, IHCI 2021, Kent, OH, USA, December 20--22, 2021, Revised Selected Papers}, pages = {689--702}, selected = {true}, year = {2021}, organization = {Springer International Publishing Cham} }