facebook

twitter

youtube

Assistant Professor & Coordinator (Students Affairs)

viswanath.g@iiitb.ac.in

Education : Ph.D. (Nanyang Technological University Singapore)

Dr. Viswanath Gopalakrishnan  finished his B.Tech from National Institute of Technology, Calicut and received Ph.D. from School of Computer Engineering, Nanyang Technological University, Singapore  in 2011. 

His Ph.D. topic was focused on building computational models for identifying salient  regions in images and videos. From 2011 - 2019 he worked with Samsung Research Institute, Bangalore on various research topics of interest  in the areas Computer Vision and Machine Learning. During this period he was involved in developing and leading the efforts  towards machine learning based vision algorithms for  Samsung's flagship models which involved Object Detection, Scene Recognition, Semantic Role Recognition and Object & Scene Segmentation. His current research areas of interest  is in Dense Scene Descriptions and Few-Shot learning techniques based on meta-learning approaches.

Computer Vision Problems , Machine learning, artificial intelligence

Major Conferences:

Random walks on graphs to model saliency in images. IEEE CVPR 2009

Sustained Observability for Salient Motion Detection. ACCV 2010, 

Unsupervised Feature Selection for Salient Object Detection, ACCV 2010

Unsupervised Temporal Segmentation of Talking Faces Using Visual Cues to Improve Emotion Recognition. Affective Computing and Intelligent Interaction (ACII), Anshul Sharma

Interactive object segmentation using single touch. International Conference on Image Processing (ICIP) ,, Kiran N Iyer

Similarity and rigidity preserving image retargeting. International Conference on Image Processing (ICIP) 2016, Anshuman Gaurav

Real-time Video Summarization on Mobile .  International Conference on Image Processing (ICIP) 2016,  

Learning rotation invariance in deep hierarchies using circular symmetric filters, ICASSP 2017, , Kiran N Iyer

Fast human segmentation using color and depth, ICASSP 2017, , Kiran N Iyer

Journals:

Salient Region Detection by Modeling Distributions of Color and Orientation, IEEE Transactions on Multimedia, 2009, 

Random Walks on Graphs for Salient Object Detection in Images. IEEE Transactions on Image Processing, 2010, 

A Linear Dynamical System Framework for Salient Motion Detection.  IEEE Trans. Circuits Syt. Video Techn, 2012 

Neutral Face Classification Using Personalized Appearance Models for Fast and Robust Emotion Detection, IEEE Transactions on Image Processing, 2015, 

Book Chapters:

Modeling Visual Saliency in Images and Videos, Computer Vision for Multimedia Applications, Methods and Solutions, IGI Global,2011,  ,Viswanath Go,  

 

Patents:

Method and apparatus for facilitating interaction with virtual reality equipment, US Patent No. 10521009, 2019, Samsung Electronics

Inventors: Kiran Nanjunda Iyer, Viswanath Gopalakrishnan, Kapil Soni, Satish Lokkoju, Sam

Method for retrieving image and electronic device thereof, US Patent No 9886762, 2018,Samsung Electronics

Inventors: Kiran Nanjunda Iyer, Raushan Kumar, Satish Lokkoju, Viswanath Gopalakrishnan, Viswanath Veera

Method and apparatus for detecting talking segments in a video sequence using visual cues, US Patent No.  9110501, 2015, Samsung Electronics.

Inventors: Sudha Velusamy, Viswanath Gopalakrishnan, Bilva Bhalachandra Navathe, Anshul Sharma

Awards:

1.) Zinnov Award 2019 for Technical Role Model in  AI and Big Data Analytics ( from 300+ nominatons across 150+ Indian Companies across 8 categories)

2.) Samsung Global CTO Award for Best Project (On-Device  Vision Algorithm Development for Galaxy S9) in Mobile Developer Conference , Suwon, Korea 2017

3.) Samsung Global CTO Award for Best Project (On-Device  Segmentation Algorithm Development ) in Mobile Developer Conference, Suwon, Korea, 2019