Akshay Krishnan

prof_pic.jpg

Hi! I am a PhD student at Georgia Tech working on 3D computer vision, advised by Prof. James Hays. I am also a student researcher at Google Research. I believe that understanding the 3D world from visual data is key to both generalizable autonomous systems and virtual reality. I enjoy working on this problem at the intersection of geometry, machine learning, and optimization.

Prior to this, I worked on long-range perception problems for autonomous driving at Waymo. I earned my Master’s from Georgia Tech in 2020, during which I worked with Prof. Frank Dellaert on SfM problems using GTSAM. I also worked with the RAIL lab during the FetchIt! challenge at ICRA 2019. I spent a wonderful summer as an intern at Blue River Technology, which also taught me a bit about agriculture.

Please feel free to send me an email if you’re interested in research collaborations!

Selected Publications

  1. omninocs.gif
    OmniNOCS: A unified NOCS dataset and model for 3D lifting of 2D objects
    Akshay Krishnan ,  Abhijit Kundu ,  Kevis-Kokitsi Maninis ,  James Hays ,  and  Matthew Brown
    In European Conference on Computer Vision , 2024
    arXiv HTML Code
  2. gtsfm.gif
    Distributed Global Structure-from-Motion with a Deep Front-End
    Ayush Baid ,  John Lambert ,  Travis Driver ,  Akshay Krishnan ,  Hayk Stepanyan ,  and  Frank Dellaert
    arXiv, 2023
    arXiv Code
  3. lane.gif
    LANe: Lighting-Aware Neural Fields for Compositional Scene Synthesis
    Akshay Krishnan ,  Amit Raj ,  Xianling Zhang ,  Alexandra Carlson ,  Nathan Tseng ,  Sandhya Sridhar ,  Nikita Jaipuria ,  and  James Hays
    arXiv, 2023
    arXiv HTML
  4. fetchit.jpg
    Taking Recoveries to Task: Recovery-Driven Development for Recipe-based Robot Tasks
    Siddhartha Banerjee ,  Angel Andres Daruna ,  Cassandra Kent ,  Weiyu Liu ,  Jonathan C. Balloch ,  Abhinav Jain ,  Akshay Krishnan ,  Muhammad Asif Rana ,  Harish Ravichandar ,  Binit Shah , and 2 more authors
    In International Symposium of Robotics Research , 2020
    arXiv