CV

Education

B.Tech. in ECE, IIIT-Hyderabad, 2009
M.S. in ECE, IIIT-Hyderabad, 2011
Ph.D. in Automatic Signal and Image Processing, Inria / Université Côte d’Azur, 2016

Work experience

Oct 2021 – Apr 2023: Applied Research Scientist II
- Amazon Prime Video, Seattle, WA, USA
- Automated trailer and poster generation from long-form videos.
Jun 2021 – Oct 2021: Research Scientist
- GE Global Research, Niskayuna, NY, USA
- Generated high-fidelity synthetic training data with StyleGAN2 to meet the accuracy demands of automated robotic inspection of aircraft turbine blades where real data was scarce.
Aug 2020 – May 2021: Applied Research Scientist
- Amazon, Hyderabad, India
- Developed stop-grouping methods for last-mile delivery, combining aerial imagery and metadata for decision-making.
Nov 2019 – Jul 2020: Research Scientist
- Self-employed, Hyderabad, India
Oct 2017 – Oct 2019: Postdoctoral Research Scientist
- Facebook Reality Labs (Oculus Research), Redmond, WA, USA
- Investigated how eye tracking, biosensing, and visual semantic information can enable naturalistic human–computer interaction. Designed and ran two user studies combining eye tracking, pupillometry, a world-facing camera, and EEG to detect cognitive-state changes in naturalistic indoor settings, and applied machine learning for event detection from multi-modal time-series data. The work was well received, resulting in an internal demo and two posters.
- Supervisor: Dr. James Hillis
Jul 2013 – Dec 2016: Doctoral Researcher
- BioVision / NeuroMathComp Team, Inria, Sophia Antipolis, France
- Developed bio-inspired models bridging computational neuroscience and computer vision for optical flow estimation and motion perception.
- Thesis: Towards synergistic models of motion information processing in biological and artificial vision.
- Supervisors: Dr. Pierre Kornprobst and Dr. Guillaume Masson
Feb 2013 – Jul 2013: Research Assistant
- NeuroMathComp Team, Inria, Sophia Antipolis, France
- Developed computational models of multistable motion perception (barber-pole stimuli). Implemented CUDA-based numerical continuation routines to simulate a 3D recurrent neural-field model, uncovering oscillatory solutions that capture spatial motion integration for gratings viewed through square and rectangular apertures.
- Supervisor: Dr. Pierre Kornprobst
Jun 2012 – Dec 2012: Research Assistant
- School of Computing, National University of Singapore, Singapore
- Prototyped a light-field acquisition system built on a 5×5 GigE camera array, developing routines for acquisition, calibration, and rendering. The work characterized the sampling challenges involved in bringing camera arrays to mobile devices.
- Supervisor: Prof. Michael Brown
Aug 2011 – Mar 2012: Consultant
- GE Global Research, Bangalore, India
- Designed and developed an automatic, reliable, high-speed pose-correction solution for 3D CT head scans, driven by key anatomical landmarks.
May 2009 – Aug 2009: Research Assistant
- Center for Visual Information Technology, IIIT-Hyderabad, India
- Developed methods for automatic image-quality estimation and restoration of color retinal fundus images.
May 2008 – Jul 2008: Intern
- GE Global Research, Bangalore, India
- Designed and developed a tool for automatic anatomical labeling and interactive visualization of neural vasculature in 3D CT head scans — recognized by the manager for delivering it within two and a half months.

Publications

Synthetic zooming of tomographic images by combination of lattices

N. Dixit, N. V. Kartheek Medathati, J. Sivaswamy. *Synthetic zooming of tomographic images by combination of lattices.* 2009 IEEE NSS/MIC, pp. 3770–3776. doi:10.1109/NSSMIC.2009.5401886

Robust matching of multi-modal retinal images using radon transform based local descriptor

Yogesh Babu Bathina, N. V. Kartheek Medathati, Jayanthi Sivaswamy. *Robust matching of multi-modal retinal images using radon transform based local descriptor.* IHI '10, ACM, 2010, pp. 765–770. doi:10.1145/1882992.1883108

Local descriptor based on texture of projections

N. V. Kartheek Medathati, Jayanthi Sivaswamy. *Local descriptor based on texture of projections.* ICVGIP '10, ACM, 2010, pp. 398–404. doi:10.1145/1924559.1924612

Detection and discrimination of disease-related abnormalities based on learning normal cases

K. Sai Deepak, N. V. Kartheek Medathati, Jayanthi Sivaswamy. *Detection and discrimination of disease-related abnormalities based on learning normal cases.* Pattern Recognition, 45(10):3707–3716, 2012. doi:10.1016/j.patcog.2012.03.020

Decoding MT motion response for optical flow estimation: An experimental evaluation

M. Chessa, N. V. Kartheek Medathati, G. S. Masson, F. Solari, P. Kornprobst. *Decoding MT motion response for optical flow estimation: An experimental evaluation.* 23rd EUSIPCO, 2015, pp. 2241–2245. doi:10.1109/EUSIPCO.2015.7362783

What can we expect from a V1-MT feedforward architecture for optical flow estimation?

Fabio Solari, Manuela Chessa, N. V. Kartheek Medathati, Pierre Kornprobst. *What can we expect from a V1-MT feedforward architecture for optical flow estimation?* Signal Processing: Image Communication, 39(B):342–354, 2015. doi:10.1016/j.image.2015.04.006

Bio-inspired computer vision: Towards a synergistic approach of artificial and biological vision

N. V. Kartheek Medathati, Heiko Neumann, Guillaume S. Masson, Pierre Kornprobst. *Bio-inspired computer vision: Towards a synergistic approach of artificial and biological vision.* Computer Vision and Image Understanding, Volume 150, 2016, Pages 1–30. DOI: 10.1016/j.cviu.2016.04.009

Recurrent network dynamics reconciles visual motion segmentation and integration.

N. V. K. Medathati, J. Rankin, A. I. Meso, P. Kornprobst, G. S. Masson. *Recurrent network dynamics reconciles visual motion segmentation and integration.* Scientific Reports 7, 11270 (2017). doi:10.1038/s41598-017-11373-z

Towards inferring cognitive state changes from pupil size variations in real world conditions

Naga Venkata Kartheek Medathati, Ruta Desai, James Hillis. *Towards inferring cognitive state changes from pupil size variations in real world conditions.* ETRA '20 Full Papers, ACM, 2020, Article 22, 1–10. doi:10.1145/3379155.3391319

Service and leadership

Program Committee member, Mutual Benefits of Cognitive and Computer Vision workshop, ICCV 2017
Student Volunteer, EUSIPCO 2015