Haithem Turki

I am a senior research scientist at NVIDIA working in the Spatial Intelligence Lab led by Professor Sanja Fidler. I am broadly interested in 3D deep learning — most of my recent work covers neural reconstruction and data-driven simulation. We're currently looking for interns for 2026, so please reach out if you are interested!

I was previously a PhD student in Computer Science at Carnegie Mellon University where I had the privilege of being advised by Professor Deva Ramanan. During that time, I collaborated with Argo AI and Reality Labs Research where I was hosted by Drs. Christian Richardt and Michael Zollhoefer. Before my PhD studies, I spent seven years at Palantir Technologies in various leadership roles across business and product development. Even earlier, I received my BS and MS in Computer Science from Stanford University where I worked with Professors Alex Aiken and Daphne Koller.

Email  /  CV  /  Google Scholar  /  Twitter  /  GitHub  /  Pro Bono

profile photo

Publications

   show selected  /  show all by date
project image

DiffusionHarmonizer: Bridging Neural Reconstruction and Photorealistic Simulation with Online Diffusion Enhancer


Yuxuan Zhang*, Katarina Tothova*, Zian Wang, Kangxue Yin, Haithem Turki, Riccardo de Lutio, Yen-Yu Chang, Or Litany, Sanja Fidler, Zan Gojcic
CVPR 2026
project page / paper

project image

SimULi: Real-Time LiDAR and Camera Simulation with Unscented Transforms


Haithem Turki*, Qi Wu*, Xin Kang, Janick Martinez Esturo, Shengyu Huang, Ruilong Li, Zan Gojcic, Riccardo de Lutio
ICLR 2026
project page / paper

project image

Lyra: Generative 3D Scene Reconstruction via Video Diffusion Model Self-Distillation


Sherwin Bahmani, Tianchang Shen, Jiawei Ren, Jiahui Huang, Yifeng Jiang, Haithem Turki, Andrea Tagliasacchi, David B. Lindell, Zan Gojcic, Sanja Fidler, Huan Ling, Jun Gao*, Xuanchi Ren*
ICLR 2026
project page / paper / code

project image

Difix3D+: Improving 3D Reconstructions with Single-Step Diffusion Models


Jay Zhangjie Wu*, Yuxuan Zhang*, Haithem Turki, Xuanchi Ren, Jun Gao, Mike Zheng Shou, Sanja Fidler, Zan Gojcic†, Huan Ling†
CVPR 2025 (Best Paper Award Candidate)
project page / paper / code

project image

HybridNeRF: Efficient Neural Rendering via Adaptive Volumetric Surfaces


Haithem Turki, Vasu Agrawal, Samuel Rota Bulò, Lorenzo Porzi, Peter Kontschieder, Deva Ramanan, Michael Zollhöfer, Christian Richardt
CVPR 2024 (Highlight)
project page / paper

project image

SpecNeRF: Gaussian Directional Encoding for Specular Reflections


Li Ma, Vasu Agrawal, Haithem Turki, Changil Kim, Chen Gao, Pedro V. Sander, Michael Zollhöfer, Christian Richardt
CVPR 2024 (Highlight)
project page / paper

project image

PyNeRF: Pyramidal Neural Radiance Fields


Haithem Turki, Michael Zollhöfer, Christian Richardt, Deva Ramanan
NeurIPS 2023
project page / paper / code

project image

Low-Bandwidth Self-Improving Transmission of Rare Training Data


Shilpa George, Haithem Turki, Ziqiang Feng, Deva Ramanan, Padmanabhan Pillai, Mahadev Satyanarayanan
MobiCom 2023
paper

project image

SUDS: Scalable Urban Dynamic Scenes


Haithem Turki, Jason Y. Zhang, Francesco Ferroni, Deva Ramanan
CVPR 2023
project page / paper / code

project image

Accelerating Silent Witness Storage


Mahadev Satyanarayanan, Ziqiang Feng, Shilpa George, Jan Harkes, Roger Iyengar, Haithem Turki, Padmanabhan Pillai
IEEE Micro, November 2022
paper

project image

Mega-NeRF: Scalable Construction of Large-Scale NeRFs for Virtual Fly-Throughs


Haithem Turki, Deva Ramanan, Mahadev Satyanarayanan
CVPR 2022
project page / paper / code

project image

OpenRTiST: End-to-End Benchmarking for Edge Computing


Shilpa George, Thomas Eiszler, Roger Iyengar, Haithem Turki, Ziqiang Feng, Junjue Wang, Padmanabhan Pillai, Mahadev Satyanarayanan
IEEE Pervasive Computing, October 2020
paper / code

project image

Edge Computing for Legacy Applications


Mahadev Satyanarayanan, Thomas Eiszler, Jan Harkes, Haithem Turki, Ziqiang Feng
IEEE Pervasive Computing, October 2020
paper

project image

Parameter Estimation and Energy Minimization for Region-Based Semantic Segmentation


M. Pawan Kumar, Haithem Turki, Dan Preston and Daphne Koller
TPAMI 2015
paper

project image

Learning Specific-Class Segmentation from Diverse Data


M. Pawan Kumar, Haithem Turki, Dan Preston and Daphne Koller
ICCV 2011
paper

Open Mentoring

A lot of good advice in research is still passed informally through networks. That can make research directions, career transitions, and even basic next steps feel unnecessarily opaque. I try to reduce that gap by setting aside a small amount of time for open mentoring conversations.

Following the example set by Wei-Chiu Ma, Jun Gao, and others, I reserve 1–2 hours most weeks for pro bono mentoring conversations. These office hours are intended for students and early-career researchers who would benefit from candid, practical advice.

I especially encourage requests from:

  • people from backgrounds underrepresented in computing,
  • people transitioning from industry into research or academia,
  • people coming from non-traditional research paths who may have less access to informal advising networks.

Conversations are informal and usually 20–30 minutes. I may not be able to provide detailed feedback on full papers, codebases, or application materials in advance.

If a conversation would be helpful, please send me a brief email with a few lines about your background, goals, and the main question you'd like to discuss.