Raja Kumar

I am a PhD student at University of Southern California advised by Prof. Ram Nevatia. I am broadly interested in Computer Vision and Deep Learning, with a focus on developing models for visual understanding in images and videos.

Before Joining USC, I completed my MS in Computer Science at the University of California Santa Cruz advised by Prof. James Davis. Previously, at Samsung Research, I worked with the on-device-AI team on developing and deploying Deep Learning Models on Samsung devices. I obtained my Bachelor's degree in Electrical Engineering from IIT Kharagpur.

Email  /  GitHub  /  Google Scholar  /  LinkedIn  /  Resume (1 Page)

profile photo

News/Updates
  • [Aug 2024] Joined USC as a CS PhD Student!
  • [Aug 2024] My work on few-shot 3D Gaussian splatting got accepted at S3DSGR @ ECCV'24!
  • [Apr 2024] Joined Flawless AI as Applied Science Intern 3D vision
  • [Mar 2024] Completed my MS in CS at UC Santa Cruz
  • [Oct 2023] Attended ICCV 2023 in Paris
  • [Aug 2023] My Paper on 3D face reconstruction got accepted at ICCV workshops 2023
  • [Sep 2022] Joined UCSC as CSE Master's student starting in Fall 2022
  • [Mar 2022] Joined Silverlabs recommendation team as an ML Engineer
  • [Apr 2021] My paper on quantization got accepted by IJCNN 2021
  • [Jun 2019] Joined Samsung Research on-device-AI team as an ML Engineer

Projects/Papers

Paper: Few-shot Novel View Synthesis using Depth Aware 3D Gaussian Splatting
Raja Kumar, Vanshika Vats
ECCV Workshops 2024

Pdf / code

Paper: Disjoint Pose and Shape for 3D Face Reconstruction
Raja Kumar, Jiahao Luo, Alex Pang, James Davis
IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, 2023


Paper: 3D Face Reconstruction: Is model-based classic passive stereo competitive?
Jiahao Luo, Raja Kumar, Alex Pang, James Davis
Under Review

Pdf

Project: Dense 2D facial landmarks detector using 3D face model
Raja Kumar


Project: Task Oriented Conversational Modelling With Subjective Knowledge
Raja Kumar

pdf / code

Paper: Hybrid and non-uniform quantization methods using retro synthesis data for efficient inference
Raja Kumar, Tejpratap GVSL, Pradeep NS,
IJCNN, 2021

pdf / code

Project: Automatic Classification and Localisation of Defects in Hot-Rolled Steel Surfaces
Raja Kumar



Teaching Assistant
  • Introduction to Artificial Intelligence, Prof. Mohammad Reza Rajati, USC, Fall 2024
  • Computer Systems and Assembly Programming, Prof. Marcelo Siero, UCSC, Winter, Fall 2023, Winter 2024
  • Introduction to Analysis of Algorithms, Prof. Suresh Lodha, UCSC, Summer 2023
  • Computer Architecture Prof. Scott Beamer, UCSC,Spring 2023


Design and source code from Jon Barron's website