1 code implementation • ECCV 2020 • Maunil R Vyas, Hemanth Venkateswara, Sethuraman Panchanathan
The SR-loss guides the LsrGAN to generate visual features that mirror the semantic relationships between seen and unseen classes.
no code implementations • 1 Jul 2019 • Piyush Papreja, Hemanth Venkateswara, Sethuraman Panchanathan
Playlists have become a significant part of our listening experience because of the digital cloud-based services such as Spotify, Pandora, Apple Music.
no code implementations • 23 Jun 2017 • Hemanth Venkateswara, Vineeth N. Balasubramanian, Prasanth Lade, Sethuraman Panchanathan
The emergence of depth imaging technologies like the Microsoft Kinect has renewed interest in computational methods for gesture classification based on videos.
no code implementations • 23 Jun 2017 • Hemanth Venkateswara, Shayok Chakraborty, Troy McDaniel, Sethuraman Panchanathan
To determine the parameters in the NET model (and in other unsupervised domain adaptation models), we introduce a validation procedure by sampling source data points that are similar in distribution to the target data.
no code implementations • 23 Jun 2017 • Hemanth Venkateswara, Prasanth Lade, Binbin Lin, Jieping Ye, Sethuraman Panchanathan
Estimating the MI for a subset of features is often intractable.
no code implementations • 22 Jun 2017 • Hemanth Venkateswara, Shayok Chakraborty, Sethuraman Panchanathan
The problem of domain adaptation (DA) deals with adapting classifier models trained on one data distribution to different data distributions.
7 code implementations • CVPR 2017 • Hemanth Venkateswara, Jose Eusebio, Shayok Chakraborty, Sethuraman Panchanathan
Domain adaptation or transfer learning algorithms address this challenge by leveraging labeled data in a different, but related source domain, to develop a model for the target domain.
no code implementations • 22 Jun 2017 • Hemanth Venkateswara, Prasanth Lade, Jieping Ye, Sethuraman Panchanathan
Popular domain adaptation (DA) techniques learn a classifier for the target domain by sampling relevant data points from the source and combining it with the target data.
1 code implementation • 2 May 2017 • Ragav Venkatesan, Hemanth Venkateswara, Sethuraman Panchanathan, Baoxin Li
Using an implementation based on deep neural networks, we demonstrate that phantom sampling dramatically avoids catastrophic forgetting.
no code implementations • NeurIPS 2011 • Qian Sun, Rita Chattopadhyay, Sethuraman Panchanathan, Jieping Ye
In this paper we propose a two-stage domain adaptation methodology which combines weighted data from multiple sources based on marginal probability differences (first stage) as well as conditional probability differences (second stage), with the target domain data.