Vanshika Vats
Computer Vision | Deep Learning | Vision for Autonomous Vehicles
I am a second-year Computer Science and Engineering Ph.D. student with Visualization and Interactive Systems (VIS) group at the University of California, Santa Cruz, advised by Prof. James Davis.
My research interests lie in Computer Vision and Deep Learning, with a primary focus on Vision for Autonomous Vehicles.
Prior to this, I was working as a Research Scientist with Mercedes-Benz Research and Development, India, broadly on motion forecasting and deep learning-based evaluation framework for perception and fusion algorithms.
News/Updates
[Apr 2024] Excited to have been awarded the AnitaB.org Advancing Inclusion Scholarship to attend GHC'24!
[Aug 2023] My first-author work "Adversarial Examples with Specular Highlights" got accepted at AROW @ ICCV'23.
[Sept 2022] Joined UCSC as a CSE Ph.D. student advised by Prof. James Davis!
[Jul 2022] My first-author work "Early Prediction of Hemodynamic Shock in Pediatric Intensive Care Units With Deep Learning on Thermal Videos" with TAVLAB-IIITD has been published in Frontiers in Physiology.
[Apr 2022] Innovation Contribution Recognition by Mercedes-Benz R&D India
[Jan 2022] Awarded Google AI Research Week Scholar-2022 being among 150 candidates selected from India and Singapore
[Dec 2021] Awarded Rising Star Alumnus Award by IGDTUW for recognition of outstanding achievement in the career
[Oct 2020] Joined Mercedes-Benz R&D India as a Research Scientist
[Sept 2020] Graduated from IIIT-Delhi with a Master's in ECE
Publications
A Survey on Human-AI Teaming with Large Pre-Trained Models
Vanshika Vats, Marzia Binta Nizam, Minghao Liu, et al.
arXiv preprint arXiv:2403.04931
2024 [Link]Assessing the Impact of Prompting Methods on ChatGPT's Mathematical Capabilities
Yuhao Chen, Chloe Wong, Hanwen Yang, ..., Vanshika Vats, James Davis
arXiv preprint arXiv:2312.15006
2024 [Link]Adversarial Examples with Specular Highlights
Vanshika Vats, Koteswar Rao Jerripothula
IEEE/CVF International Conference on Computer Vision Workshops (ICCVw)
2023 [Link]Early Prediction of Hemodynamic Shock in Pediatric Intensive Care Units With Deep Learning on Thermal Videos
Vanshika Vats, Aditya Nagori, Pradeep Singh, Raman Dutt, Harsh Bandhey, Mahika Wason, Rakesh Lodha, Tavpritesh Sethi
Frontiers in Physiology, 1284
2022 [Link | PDF]A Prospectively Validated Generalizable Model for Outcome Prognostication Using Shock Index in Intensive Care Units
Aditya Nagori, Pradeep Singh, Sameena Firdos, Vanshika Vats, Arushi Gupta, Harsh Bandhey, Anushtha Kalia, Arjun Sharma, Prakriti Ailavadi, Raghav Awasthi, Wrik Bhadra, Ayushmaan Kaul, Rakesh Lodha, Tavpritesh Sethi
2021 [Link]SURF-SVM Based Identification and Classification of Gastrointestinal Diseases in Wireless Capsule Endoscopy
Vanshika Vats, Pooja Goel, Amodini Agarwal, Nidhi Goel
IEEE-ICSPVCE 2019
2019 [Link]
Research Experience
Graduate Student Researcher, VIS Group, UC Santa Cruz, CA
Sept 2022 - Present
Working with Visualization and Interactive Systems (VIS) group on advanced sensing technology for autonomous vehicles
Exploring 3D object detection on the LiDAR point clouds using Point-Voxel feature set on the KITTI dataset with an aim to fuse it with the camera 2D image modality for better perception and hence, navigation
Research Scientist, Mercedes-Benz R&D, India
Oct 2020 - July 2022
Motion forecasting through Graph Neural Networks (GNNs)
Responsible for building and scaling an end-to-end framework to assess deep learning based perception and fusion algorithms for large scale drive data at Level-4 automation
Computer Vision Researcher, NeatAI ServoLab, IIIT Delhi
Aug 2021 - April 2022
Examined natural adversaries affecting the deep neural vision models noting a ∼35% drop in the model performance. Found significant shifts in model attention after introducing the adversaries leading to exploring solutions [Manuscript under review]
Researcher, TavLab, IIIT Delhi
Jan 2021 - May 2021
Worked on elevating the limited generalization of Intensive Care Unit (ICU) vitals in predicting hemodynamic shock by bridging the gap across age-groups, ICU settings, geographies, and non-real-time settings by Deep Learning methods
Achieved AUPRC > 90% across each considered ICU site along with identifying 92% of all shock events up to more than 8 hours in advance
Education
Doctor of Philosophy, 2022 - Present
University of California, Santa Cruz | Computer Science and Engineering
Master of Technology, 2018 - 2020
Indraprastha Institute of Information Technology, Delhi | Electronics and Communication Engineering
Bachelor of Technology, 2014 - 2018
Indira Gandhi Delhi Technical University for Women, Delhi | Electronics and Communication Engineering
Hobbies
Outside of my professional life, I actively pursue my passion for photography and continually work to improve my skills and techniques. I also enjoy exploring the great outdoors.
[Instagram: @vanshikavats.nef]
