Aakash KT

Ph.D candidate at IIIT-H

profile_pic.jpg

aakash.kt@research.iiit.ac.in

I am a second year Ph.D candidate at CVIT, IIIT Hyderabad, supervised by Dr. P. J. Narayanan (Funded by the KCIS fellowship).
Previously, I completed my MS by research at IIIT Hyderabad, which is also where I completed my BTech in Computer Science Engineering.
My research involves playing around with the light transport equation for real-time performance, differentiable rendering and neural rendering (in other words, I work on Computer Vision/Computer Graphics and Machine Learning/Deep Learning.).

Welcome to my page!

Resume / Email / Google Scholar / Twitter


News


May 5, 2022 We received the best paper award at I3D for our paper “Bringing Linearly Transformed Cosines to Anisotropic GGX”! NVIDIA was generous to gift us a RTX 3090 for this. Thank you!
Mar 29, 2022 I will be joining Meta Reality Labs, Pittsburgh (formerly Facebook Reality Labs) as a Research Scientist Intern in June. I will be working in Yaser Sheikh’s team, with Giljoo Nam. Looking forward to some exciting work on Photorealistic Telepresence!
Mar 10, 2022 Our paper “Bringing Linearly Transformed Cosines to Anisotropic GGX” has been accepted to I3D 2022! This work was in collaboration with Eric Heitz and Jonathan Dupuy from Unity Research Grenoble.
Mar 4, 2022 Our work “Precomputed Radiance Transfer Using Transfer Textures” has been accepted as a poster to EG 2022.
Jun 5, 2021 Our paper “Fast Analytic Soft Shadows from Area Lights” has been accepted to EGSR 2021!
Aug 1, 2020 I have joined as a Ph.D student at CVIT, IIIT Hyderabad, supervised by Dr. P. J. Narayanan. My Ph.D is funded by the KCIS fellowship. Curious as to why I decided to do this? See here.
Feb 1, 2020 I was to join the GraphDeco research group @ Inria (Sophia Antipolis, France), as a pre-doctoral candidate from September 2020, supervised by Dr. George Drettakis and Dr. Adrien Bousseau. This was unfortunately rescinded due to COVID.

Selected Publications


  1. Bringing Linearly Transformed Cosines to Anisotropic GGX

    ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games (I3D) 2022, PACM-CGIT, Best Paper Award

    We present an extension to LTCs for anisotropic GGX, in the context of real-time analytic area light shading. Our approach robustly fits LTC matrices to anisotropic GGX and ensures artefact free rendering. The end result is a 84 LUT parameterized by the elevation and azimuth of view vector and the roughness in x and y directions.


  2. Fast Analytic Soft Shadows from Area Lights

    Eurographics Symposium on Rendering (EGSR) 2021, DL-only track

    We present an analytical solution for soft shadows from area lights, which naturally produces noise-free renderings as compared to equivalent stochastic methods. A structured approach to analytically compute soft shadows from spherical projections of lights and occluders with any 3D shape and efficiently for convex 3D shapes.


  3. A Flexible Neural Renderer for Material Visualization

    ACM SIGGRAPH Asia 2019, Technical Briefs

    Photo realism in computer generated imagery is crucially dependent on how well an artist is able to recreate real-world materials in the scene. We propose a convolutional neural network based workflow which generates high-quality ray traced material visualizations on a shaderball, in real-time.