Svati (Dhamija) Bendale

I am a PhD student in Computer Science advised by Prof.Terrance E. Boult. I work as a Research Assistant at the Vision And Security Technology (VAST)lab in University of Colorado Colorado Springs. My research lies at the intersection of affective computing, computer vision and machine learning.

Currently, I am involved in developing interactive cognitive systems that empower individuals suffering from post traumatic stress by combining sensing and context based video analysis of human behavior. As a part of this work, I am primarily focused on problems such as heart rate prediction from facial videos, engagement prediction for web-intervention and machine learning based self-efficacy methods.

I have research interests in various related areas such as facial expression analysis, facial landmark tracking, biosensor data analysis, deep learning methods for time series data, sequential data and others.


Publications

A Multimodal approach for predicting changes in PTSD symptom severity

ACM International Conference on Multimodal Interaction (ICMI), 2018, Boulder
Authors: Adria Mallol-Ragolta, Svati Dhamija and Dr. Terrance Boult
Paper


Automated Action Units Vs. Expert Raters: Face off

IEEE Winter Conference on Applications of Computer Vision (WACV), 2018, Lake Tahoe
Authors: Svati Dhamija and Dr. Terrance Boult
Paper


Learning Visual Engagement for Trauma-Recovery

Workshop on Computer Vision for Active and Assisted Living (CV-AAL), WACV 2018, Lake Tahoe
Authors: Svati Dhamija and Dr. Terrance Boult
Paper


Automated Mood-Aware Engagement Prediction

IEEE Affective Computing for Intelligent Interaction (ACII), Main Conference 2017, San Antonio
Authors: Svati Dhamija and Dr. Terrance Boult
Paper


Learning based Visual Engagement and Self-Efficacy

IEEE Affective Computing for Intelligent Interaction (ACII), Doctoral Consortium 2017, San Antonio
Author: Svati Dhamija
Paper Poster


Exploring Contextual Engagement for Trauma Recovery

Workshop on Deep Affective Learning and Context Modelling (DALCOM), CVPR 2017, Hawaii
Authors: Svati Dhamija and Dr. Terrance Boult
Paper Poster


Recurrence of Heart Rate and Skin Conductance during Web-based intervention for Trauma survivors

International Society for Traumatic Stress Studies (ISTSS), 2017
Authors: Dr.Kotaro Shoji, Amanda Devane, Svati Dhamija, Dr.Terrance Boult and Dr.Charles C.Benight


Measuring engagement into the web-intervention by the quality of voice

International Society for Research on Internet Interventions (ISRII), 8th Scientific Meeting 2016
Authors: Dr.Kotaro Shoji, Dr.Charles C.Benight, Austin Mullings, Carrie Yeager, Svati Dhamija and Dr.Terrance Boult
Poster


The importance of self-appraisals of coping capability in predicting engagement in a web intervention for trauma

International Society for Research on Internet Interventions (ISRII), 8th Scientific Meeting 2016
Authors: Dr.Charles C.Benight, Dr.Kotaro Shoji, Carrie Yeager, Austin Mullings, Svati Dhamija and Dr.Terrance Boult
Poster


Changes self-appraisal and mood utilizing a web-based recovery system on posttraumatic stress symptoms: A laboratory experiment

International Society for Traumatic Stress Studies (ISTSS), 2016, Symposium
Authors: Dr.Charles C.Benight, Dr.Kotaro Shoji, Carrie Yeager, Austin Mullings, Svati Dhamija and Dr.Terrance Boult


Comparative analysis for discrete sine transform as a suitable method for noise estimation

IJCSI International Journal of Computer Science Issues, Vol. 8, Issue 5, No 3, September 2011
Authors: Swati Dhamija and Dr. Priyanka Jain
Paper ISSN (Online): 1694-0814


Adaptive filtering algorithms for channel equalization and echo cancellation

International Journal of Computer Technology and Applications 2011
Authors: Swati Dhamija and Prof. Akash Tayal
Paper ISSN: 2229-6093


EASE Dataset

The EASE dataset has username(s)/subjects id(s) beginning with "EASE" followed by 3 numbers. For brevity, the file naming convention in the dataset uses only numbers. Engagement self-reports are provided in the Engagement_SR.xlsx with the frame number and the engagement level reported by the subject in each of the modules of Session 1 and 2. Feature files extracted by processing facial videos of EASE subjects are saved in AU_feature_files folder in text format. More details about the EASE dataset are provided in the papers. The Mood_self_reports.txt text file contains moods (POMS) data. The naming convention in the Mood_self_reports file is as follows :
(1) The filename 024_01_02_TR is data for username EASE024 -- Session1 -- Module2 -- Triggers.
(2) There are 2 sets of mood scores in the file : pre and post.
Data

If you use the dataset please cite the following publication:

@inproceedings{dhamija2017exploring,
title={Exploring contextual engagement for trauma recovery},
author={Dhamija, Svati and Boult, Terrance E},
booktitle={Computer Vision and Pattern Recognition Workshops (CVPRW), 2017 IEEE Conference on},
pages={2267--2277},
year={2017},
organization={IEEE}
}



Email: svati.dhamija@gmail.com
Curriculum vitae: PDF