Opens in a new tab

Computer graphics Generative AI

Krishna Mullia.

Research and production work in neural rendering, 3D representations, and generative models.

At Canva Research, I’m a Staff Research Engineer working on generative video systems and neural 3D.

Krishna Mullia
Canva Research
San Francisco, California

About

Before Canva, I spent seven years at Adobe Research, working on neural rendering methods and production rendering systems. Earlier, I worked on shading systems for MoonRay at DreamWorks, with film credits on Trolls, The Boss Baby, and Trolls Holiday.

I earned an M.S. in Computer Science from UC San Diego (2015), focused on computer graphics and advised by Ravi Ramamoorthi.

Selected work

Rendering systems shipped in Adobe’s creative tools.

Photoshop

Rotate Object

Rotate an object in a photograph to change its angle and perspective. My work on the feature included the custom Gaussian splatting renderer powering its interactive 3D preview.

Illustrator

3D & Materials

Ray-traced lighting and materials for vector artwork. I contributed to the renderer and co-developed the neural denoiser introduced with Illustrator’s redesigned 3D tools.

Dimension

GPU ray tracing

GPU ray tracing for Dimension, built on OptiX and RTX. I led the four-person team that took the renderer from a research prototype into production.

Substance 3D

Vulkan ray tracing & denoising

I helped bring Vulkan ray tracing and neural denoisers I co-developed to Substance 3D Stager and the then-beta Substance 3D Viewer, supporting interactive previews and faster final renders.

Selected research

Publications

Google Scholar
  1. 2026

    Stochastic Ray Tracing for the Reconstruction of 3D Gaussian Splatting

    Peiyu Xu, Shuang Zhao, Xin Sun, Krishna Mullia, Raymond Fei, Iliyan Georgiev

    CVPR 2026

  2. 2025

    Generative Neural Materials

    Nithin Raghavan*, Krishna Mullia*, Alexander Trevithick, Fujun Luan, Miloš Hašan, Ravi Ramamoorthi

    SIGGRAPH 2025 · * Equal contribution

  3. 2025

    Uncertainty for SVBRDF Acquisition using Frequency Analysis

    Ruben Wiersma, Julien Philip, Miloš Hašan, Krishna Mullia, Fujun Luan, Elmar Eisemann, Valentin Deschaintre

    SIGGRAPH 2025

  4. 2024

    RNA: Relightable Neural Assets

    Krishna Mullia, Fujun Luan, Xin Sun, Miloš Hašan

    ACM Transactions on Graphics, 2024 · Presented at SIGGRAPH 2025

  5. 2024

    Neural Product Importance Sampling via Warp Composition

    Joey Litalien, Miloš Hašan, Fujun Luan, Krishna Mullia, Iliyan Georgiev

    SIGGRAPH Asia 2024

  6. 2022

    Rendering Neural Materials on Curved Surfaces

    Alexandr Kuznetsov, Xuezheng Wang, Krishna Mullia, Fujun Luan, Zexiang Xu, Miloš Hašan, Ravi Ramamoorthi

    SIGGRAPH 2022

  7. 2021

    NeuMIP: Multi-Resolution Neural Materials

    Alexandr Kuznetsov, Krishna Mullia, Zexiang Xu, Miloš Hašan, Ravi Ramamoorthi

    ACM Transactions on Graphics · SIGGRAPH 2021

  8. 2021

    Interactive Monte Carlo Denoising using Affinity of Neural Features

    Mustafa Işık, Krishna Mullia, Matthew Fisher, Jonathan Eisenmann, Michaël Gharbi

    ACM Transactions on Graphics · SIGGRAPH 2021

Mentorship & service

Mentorship

I’ve mentored interns in rendering, neural materials, and 3D reconstruction, alongside researchers and engineers at Canva.

Interns & collaborators

Neham Jain, Elmer Vasquez, Ruben Wiersma, Joey Litalien, Nithin Raghavan, Bailey Miller, Vaibhav Vavilala, Alexandr Kuznetsov, Mustafa Işık, Yaoyi Bai, and Owen Jow.

Service

Regular reviewing for SIGGRAPH, SIGGRAPH Asia, and Eurographics; program committee service includes High-Performance Graphics, the Eurographics Symposium on Rendering, and SIGGRAPH Asia Technical Communications and Posters.

Contact

Email is the best way to reach me.