|
Ekta Prashnani Research Scientist NVIDIA [GitHub] [Google Scholar] [CV] |
I am a Research Scientist in NVIDIA's Human
Performance and Experience team.
My research is primarily focused on perceptual tasks related to evaluating the quality and authenticity of images and videos.
To facilitate my research on these topics, I am also very interested in several related research directions: motion processing in videos, learning from limited or noisy data, and generative models for images and videos.
Some past notable works include PieAPP, a perceptually-consistent image error assessment method and accompanying dataset (arXiv link), and
a noise-aware training paradigm for video saliency prediction (arXiv link).
During my free time, I enjoy running long distances, painting misty forests (for which I learn from Sarah McKendry and Fiona Dalrymple)
and taking pictures.
News
| Apr, 2022: Started as a Research Scientist in the Human Performance and Experience team at NVIDIA. |
| Feb, 2022: Successfully defended my thesis "Data-driven Methods for Evaluating the Perceptual Quality and Authenticity of Visual Media". |
| Oct, 2021: Participated in ICCV 2021 Doctoral Consortium. |
| Jun, 2021: ECE Dissertation Fellowship by the ECE department at UCSB. |
| Apr, 2021: Our work on noise-aware training strategies for visual saliency prediction (accepted at BMVC'21) is now available on arXiv. |
| Jan, 2021: The PieAPP dataset is now available publicly! See here for details. |
| Jul, 2019: Started my internship at Nvidia Research, Learning and Perception Research team. |
| May, 2019: Outstanding Teaching Assistant award by the ECE department at UCSB. |
| Jun, 2018: Our CVPR2018 paper about a new perceptual image-error metric (PieAPPv0.1) and the associated source code and trained model is now available online! |
| May, 2018: Outstanding Teaching Assistant award by the ECE department at UCSB. |
Publications
E. Prashnani, Orazio Gallo, Joohwan Kim, Josef Spjut, Pradeep Sen, Iuri Frosio, "Noise-Aware Video Saliency Prediction," British Machine Vision Conference, 2021.  |
E. Prashnani*, H. Cai*, Y. Mostofi and P. Sen, "PieAPP: Perceptual Image-Error
Assessment through Pairwise Preference," Computer Vision and Pattern Recognition, 2018.  |
| E. Prashnani, M. Moorkami, D. Vaquero and P. Sen, "A Phase-Based Approach for Animating Images Using Video Examples," Computer Graphics Forum, August 2016, Volume 36,
Issue 6.
|
| *joint first authors |
Teaching
|
Technical Mentorship for EE Capstone (2018-2019) Provided technical mentorship to seniors on their EE Capstone projects (total five capstone projects). I worked very closely with the capstone team (left to right in the top picture: Benjamin Hirt, Erik Rosten, Shan-Wei Sun) working on the design and deployment of machine-learning-based medical image recognition algorithms for arthroscopic images (sponsored by Arthrex). The team used an Nvidia Jetson TX2 interfaced with the live camera feed from an Arthrex surgical drawer, to deploy the trained deep-learning-based models. Check out a brief live demo here (starting at 25s). |
|
Technical Mentorship for EE Capstone (2017-2018) Provided technical mentorship to seniors on their EE Capstone projects (total five capstone projects). I worked very closely with the capstone team working on medical image recognition for arthroscopic images (sponsored by Arthrex). The team (left to right: Jonathan Huynh, Phanitta Chomsinsap, Jacob Kurtz and Alae Amara) was selected to present their work at the Engineering Design Expo, 2018, at UCSB. |
|
Research Mentor for High School Students (July 2017) In the Summer of 2017, I had the opportunity to mentor four exceptional high school students (left to right in first picture: Sohini Kar, Jungwoo Park, Joshua Doolan, James Wang) as a part of the Summer Research Mentorship Program at UCSB. I spent the first few weeks of the program teaching relevant concepts of computer vision and machine learning to these students (they followed along easily - the age for brilliance keeps getting younger!). The students spent the latter half of the program working on the research tasks I designed for them in applying deep learning to object detection (for Sohini and Jungwoo) and image restoration (for Joshua and James). |
|
Technical Mentorship for EE Capstone (2016-2017) Provided technical mentorship to seniors on their EE Capstone projects (total six capstone projects). I worked very closely with the capstone team working on deep-learning-based image super-resolution (sponsored by Flir). The team (left to right: Julian Castro, Connor Northend, Jose Jimenez) ended up winning the award for the Best Technical Capstone Project! |