Monde Duinkharjav

Budmonde Duinkharjav
Дуйнхаржавын Будмондэ

Profile Picture
Clicky Stuffs
Email: budmonde@nyu.edu
Twitter: @budmonde
Github: budmonde
...
if it exists, it's "budmonde".
Resume
Google Scholar
Last Update: Mar 6, 2025

I am a PhD candidate at NYU's Immersive Computing Lab, advised by Prof. Qi Sun. My research is focused around studying the relationship between what we observe in our surroundings and how its perception affects our behavior and ability to perform visual tasks in the context of comptuer graphics applications. More broadly, I'm interested in how research on human perception can be leveraged to augment and improve computer graphics systems to aide in our daily lives.

Prior to starting my studies at NYU in Spring 2021, I received my BS and MEng degrees in Computer Science and Engineering at MIT in 2018 and 2019 respectively. During my time at MIT, I was part of the Computer Graphics Group at CSAIL, and was advised by Prof. Fredo Durand, where I worked on incorporating differentiable ray tracing into machine learning pipelines.

Updates

Aug 2024
I plan to graduate in May 2025, and I'm on the market looking for full-time opportunities as a research scientist.
May 2024
I started an internship at NVIDIA's AI for Graphics and Gaming team mentored by Anjul Patney and Loudon Cohen.
May 2023
I started an internship at Adobe Research mentored by Chang Xiao.
May 2023
I received an award for Outstanding Performance on the PhD Qualification Exam at NYU.
Dec 2022
I got an Honorable Mention at the Snap Research Fellowship.
Oct 2022
A paper I contributed to received the Best Journal Paper Award at IEEE ISMAR 2022.
Aug 2022
My first-authored paper received the Best Paper Award at ACM SIGGRAPH 2022.
May 2022
I started an internship at NVIDIA's Human Performance and Experience Team (HPX) mentored by Rachel Brown.

Publications

Profile Picture
FovealNet: Advancing AI-Driven Gaze Tracking Solutions for Efficient Foveated Rendering in Virtual Reality
Wenxuan Liu, Budmonde Duinkharjav, Qi Sun, Sai Qian Zhang
IEEE VR 2025
Eye tracking prediction based on IR camera footage can be improved by implementing a transformer-based neural network model.
project page
doi
arXiv
Profile Picture
Evaluating Visual Perception of Object Motion in Dynamic Environments
Budmonde Duinkharjav, Jenna Kang, Gavin S. P. Miller, Chang Xiao, Qi Sun
SIGGRAPH Asia 2024  |  Journal
Human estimation of object movements in a 3D environment exhibits perceptual errors during self-movement; we model and predict this estimation error towards designing human perception-aware content design.
Project Page
doi
Citation duinkharjav2024motionestimation  copied to clipboard 
Profile Picture
Exploiting Human Color Discrimination for Memory- and Energy-Efficient Image Encoding in Virtual Reality
Nisarg Ujjainkar, Ethan Shahan, Kenneth Chen, Budmonde Duinkharjav, Qi Sun, Yuhao Zhu
ASPLOS 2024
A frame buffer compression scheme for HMDs is achieved by leveraging peripheral deterioration of color discrimination.
Project Page
doi
arXiv
Citation ujjainkar2024exploiting  copied to clipboard 
Profile Picture
The Shortest Route Is Not Always the Fastest: Probability-Modeled Stereoscopic Eye Movement Completion Time in VR
Budmonde Duinkharjav, Benjamin Liang, Anjul Patney, Rachel Brown, Qi Sun
SIGGRAPH Asia 2023  |  Journal
The amplitude and direction of gaze movements in binocular vision exhibit different temporal performance characteristics; our computational model predicts this relationship.
Project Page
doi
arxiv
Citation duinkharjav2023stereolatency  copied to clipboard 
Profile Picture
Color-Perception-Guided Display Power Reduction for Virtual Reality
Budmonde Duinkharjav*, Kenneth Chen*, Abhishek Tyagi, Jiayi He, Yuhao Zhu, Qi Sun(* co-authors)
SIGGRAPH Asia 2022  |  Journal
A power saver for OLED modules in untethered HMDs is achieved by applying a perceptually unnoticeable foveated color modulation filter.
Project Page
doi
arXiv
Citation duinkharjav2022vrpowersaver  copied to clipboard 
Profile Picture
Reconstructing room scales with a single sound for augmented reality displays
Benjamin Liang, Andrew Liang, Iran Roman, Tomer Weiss, Budmonde Duinkharjav, Juan Pablo Bello, Qi Sun
JID 2022
Room dimensions can be inferred from hearing how an audio signal propogates within.
Project Page
doi
Citation liang2022audioreconstruction  copied to clipboard 
Profile Picture
FoV-NeRF: Foveated Neural Radiance Fields for Virtual Reality
Nianchen Deng, Zhenyi He, Jiannan Ye, Budmonde Duinkharjav, Praneeth Chakravarthula, Xubo Yang, Qi Sun
ISMAR 2022Best Journal Paper
Neural Radiance Field performance can be accelerated while perserving visual fidelity in egocentric applications via a coordinate system re-parameterization.
project page
doi
arXiv
Citation deng2021fovnerf  copied to clipboard 
Profile Picture
Image Features Influence Reaction Time: A Learned Probabilistic Perceptual Model for Saccade Latency
Budmonde Duinkharjav, Praneeth Chakravarthula, Rachel Brown, Anjul Patney, Qi Sun
SIGGRAPH 2022  |  JournalBest Paper
Human decision making performance varies as a function of the image features observed; this relationship can be leveraged to tailor graphics applications for best performance.
Project Page
doi
arXiv
blog
Citation duinkharjav2022gazetiming  copied to clipboard 
Profile Picture
Instant Reality: Gaze-Contingent Perceptual Optimization for 3D Virtual Reality Streaming
Shaoyu Chen, Budmonde Duinkharjav, Xin Sun, Li-Yi Wei, Stefano Petrangeli, Jose Echevarria, Claudio Silva, Qi Sun
TVCG 2022
Data transmission bandwidth is saved by determining the LOD of 3D assets in a foveated manner.
Project Page
doi
arXiv
Citation chen2022irg  copied to clipboard 

Posters and Demos

Profile Picture
Imperceptible Color Modulation for Power Saving in VR/AR
Kenneth Chen, Budmonde Duinkharjav, Nisarg Ujjainkar, Ethan Shahan, Abhishek Tyagi, Jiayi He, Yuhao Zhu, Qi Sun
SIGGRAPH 2023  |  Emerging Technologies
Follow-up live demo of the power saver for OLED modules in untethered HMDs; see Color-Perception-Guided Display Power Reduction for Virtual Reality for details.
Project Page
doi
Citation chen2023vrpowersaver  copied to clipboard 
Profile Picture
Learning Non-stationary SVBRDFs using GANs and Differentiable Rendering
Budmonde Duinkharjav
MIT MEng Thesis 2019
Texture-maps for 3D assets can be learned directly from rendered images via differentiable ray tracing and GANs.
Paper
Citation duinkharjav2019svbrdfgan  copied to clipboard 

Teaching Experience

MIT 6.815/865: Digital and Computational Photography: Teaching Assistant, Spring 2019

MIT 6.858: Computer Systems Security: Teaching Assistant, Spring 2018

MIT 6.148: WebLab: Introduction to Web Programming: Co-Instructor
Lecture Videos: Winter 2017 and Winter 2018