Publications | Aakash KT

2023

Combining Resampled Importance and Projected Solid Angle Samplings for Many Area Light Rendering

Ishaan Shah*, Aakash KT*, and P. J. Narayanan

ACM SIGGRAPH Asia 2023, Technical Communications (Formerly Technical Briefs)

[PDF] [Code] [Project Page] [Genesis]

We identify the core issue for the high run-times of a naive combination of RIS and projected solid angle sampling. We then reformulate RIS for a better integration with projected solid angle sampling, achieving the state-of-the-art in many area light rendering.
Analytical & Neural approaches to Physically Based Rendering

Aakash KT

ACM SIGGRAPH Asia 2023, Doctoral Consortium

[PDF]

Summary of my PhD thesis, including two extensions to LTCs and the extension of NRC for efficient hair rendering.

Accelerating Hair Rendering by Learning High-Order Scattered Radiance

Aakash KT, Adrian Jarabo, Carlos Aliaga, Matt Jen-Yuan Chiang, Olivier Maury, Christophe Hery, P. J. Narayanan, and Giljoo Nam

Eurographics Symposium on Rendering (EGSR) 2023, Computer Graphics Forum (CGF) Vol. 42, No. 4

[BibTex] [PDF] [Code] [Project Page] [Genesis]

Efficiently and accurately rendering hair accounting for multiple scattering is a challenging open problem. We present a technique to infer the higher order scattering in hair in constant time within the path tracing framework, while achieving better computational efficiency.

@article{10.1111:cgf.14895,
  journal = {Computer Graphics Forum},
  title = {{Accelerating Hair Rendering by Learning High-Order Scattered Radiance}},
  author = {KT, Aakash and Jarabo, Adrian and Aliaga, Carlos and Chiang, Matt Jen-Yuan and Maury, Olivier and Hery, Christophe and Narayanan, P. J. and Nam, Giljoo},
  year = {2023},
  publisher = {The Eurographics Association and John Wiley & Sons Ltd.},
  issn = {1467-8659},
  doi = {10.1111/cgf.14895},
  conference_display_name = {Eurographics Symposium on Rendering (EGSR) 2023},
  track_display_name = {Computer Graphics Forum (CGF) Vol. 42, No. 4},
  icon = {/assets/img/egsr2023.png},
  genesis = {/hair_high_order_genesis}
}

2022

Real-Time Rendering of Arbitrary Surface Geometries using Learnt Transfer

Sirikonda Dhawal, Aakash KT, and P. J. Narayanan

ICVGIP 2022, Full Paper

[BibTex] [PDF] [Project Page]

In this paper, we propose a compact transfer representation that is learnt directly on scene geometry points. Specifically, we train a small multi-layer perceptron (MLP) to predict the transfer at sampled surface points.

@inproceedings{dhawal2022prtsdf,
  author = {Dhawal, Sirikonda and KT, Aakash and Narayanan, P. J.},
  title = {Real-Time Rendering of Arbitrary Surface Geometries using Learnt Transfer},
  year = {2022},
  doi = {10.1145/3571600.3571640},
  booktitle = {Proceedings of the Thirteenth Indian Conference on Computer Vision, Graphics and Image Processing},
  conference = {ICVGIP 2022 - (To appear in ACM Digital Library)},
  conference_display_name = {ICVGIP 2022},
  track_display_name = {Full Paper},
  icon = {/assets/img/icvgip_2022_dhawal.png}
}

Learnt Transfer for Surface Geometries

Sirikonda Dhawal, Aakash KT, and P. J. Narayanan

High Performance Graphics 2022 (HPG), Poster

[BibTex] [PDF] [Project Page]

@inproceedings{dhawal_learnt_prt,
  author = {Dhawal, Sirikonda and KT, Aakash and Narayanan, P. J.},
  title = {Learnt Transfer for Surface Geometries},
  year = {2022},
  conference_display_name = {High Performance Graphics 2022 (HPG)},
  track_display_name = {Poster},
  icon = {/assets/img/mlp_fit.png},
}

Bringing Linearly Transformed Cosines to Anisotropic GGX

Aakash KT, Eric Heitz, Jonathan Dupuy, and P. J. Narayanan

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

[BibTex] [PDF] [Supp] [Code] [Slides] [Project Page] [Genesis]

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.

@article{10.1145/3522612,
  author = {KT, Aakash and Heitz, Eric and Dupuy, Jonathan and Narayanan, P. J.},
  title = {Bringing Linearly Transformed Cosines to Anisotropic GGX},
  year = {2022},
  issue_date = {May 2022},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  volume = {5},
  number = {1},
  url = {https://doi.org/10.1145/3522612},
  doi = {10.1145/3522612},
  journal = {Proc. ACM Comput. Graph. Interact. Tech.},
  month = may,
  articleno = {12},
  numpages = {18},
  keywords = {area lighting, BRDF, LTC, real-time rendering},
  conference_display_name = {ACM SIGGRAPH Symposium on Interactive 3D Graphics and Games (I3D) 2022},
  track_display_name = {PACM-CGIT},
  best_paper = {true},
  icon = {/assets/img/i3d2022.jpg},
  genesis = {/aniso_ltc_genesis}
}

PRTT: Precomputed Radiance Transfer Textures

Sirikonda Dhawal, Aakash KT, and P. J. Narayanan

Eurographics (EG) 2022, Poster

[BibTex] [PDF] [Poster]

We analyze and extend PRT to use textures for storing transfer, instead of at vertices of a mesh. We demonstrate better rendering quality for the same mesh resolution for glossy reflection and inter-reflections. We also analyze the run-time, memory requirements and demonstrate benefits of using transfer textures.
@article{dhawal2022prtt, title = {PRTT: Precomputed Radiance Transfer Textures}, author = {Dhawal, Sirikonda and KT, Aakash and Narayanan, P. J.}, journal = {arXiv preprint arXiv:2203.12399}, year = {2022}, conference_display_name = {Eurographics (EG) 2022}, track_display_name = {Poster}, icon = {/assets/img/prt_transfer_tex.jpg}, }

2021

Fast Analytic Soft Shadows from Area Lights

Aakash KT, Parikshit Sakurikar, and P. J. Narayanan

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

[BibTex] [PDF] [Code] [Slides] [Genesis]

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.

@inproceedings{.20211295,
  booktitle = {Eurographics Symposium on Rendering - DL-only Track},
  editor = {Bousseau, Adrien and McGuire, Morgan},
  title = {{Fast Analytic Soft Shadows from Area Lights}},
  author = {KT, Aakash and Sakurikar, Parikshit and Narayanan, P. J.},
  year = {2021},
  publisher = {The Eurographics Association},
  issn = {1727-3463},
  isbn = {978-3-03868-157-1},
  doi = {10.2312/sr.20211295},
  conference_display_name = {Eurographics Symposium on Rendering (EGSR) 2021},
  track_display_name = {DL-only track},
  icon = {/assets/img/egsr2021.png},
  genesis = {/fast_ss_genesis}
}

Neural View Synthesis with Appearance Editing from Unstructured Images

Pulkit Gera, Aakash KT, Sirikonda Dhawal, and P. J. Narayanan

ICVGIP 2021, Full Paper

[BibTex] [PDF] [Code] [Project Page]

We present a neural rendering framework for simultaneous view synthesis and appearance editing of a scene with known environmental illumination captured using a mobile camera. Our approach explicitly disentangles the appearance and learns a lighting representation that is independent of it. We show results of editing the appearance of real scenes in interesting and non-trivial ways.

@inproceedings{10.1145/3490035.3490299,
  author = {Gera, Pulkit and KT, Aakash and Dhawal, Sirikonda and Narayanan, P. J.},
  title = {Neural View Synthesis with Appearance Editing from Unstructured Images},
  year = {2021},
  isbn = {9781450375962},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  url = {https://doi.org/10.1145/3490035.3490299},
  doi = {10.1145/3490035.3490299},
  booktitle = {Proceedings of the Twelfth Indian Conference on Computer Vision, Graphics and Image Processing},
  articleno = {40},
  numpages = {9},
  keywords = {text tagging, LATEX, ACM proceedings},
  location = {Jodhpur, India},
  series = {ICVGIP '21},
  conference_display_name = {ICVGIP 2021},
  track_display_name = {Full Paper},
  icon = {/assets/img/icvgip.jpg},
}

2019

Exploring Data Driven Graphics for Interactivity

Aakash KT, and P. J. Narayanan

IIIT-H, Master’s Thesis

[PDF]

A Flexible Neural Renderer for Material Visualization

Aakash KT, Parikshit Sakurikar, Saurabh Saini, and P. J. Narayanan

ACM SIGGRAPH Asia 2019, Technical Briefs

[arXiv] [BibTex] [PDF] [Code] [Project Page] [Genesis]

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.

@inproceedings{KT:2019:FNR:3355088.3365160,
  author = {KT, Aakash and Sakurikar, Parikshit and Saini, Saurabh and Narayanan, P. J.},
  title = {A Flexible Neural Renderer for Material Visualization},
  booktitle = {SIGGRAPH Asia 2019 Technical Briefs},
  series = {SA '19},
  year = {2019},
  isbn = {978-1-4503-6945-9},
  location = {Brisbane, QLD, Australia},
  pages = {83--86},
  numpages = {4},
  url = {http://doi.acm.org/10.1145/3355088.3365160},
  doi = {10.1145/3355088.3365160},
  acmid = {3365160},
  publisher = {ACM},
  address = {New York, NY, USA},
  keywords = {Deep Learning, Global Illumination, Neural Rendering, Ray Tracing},
  conference_display_name = {ACM SIGGRAPH Asia 2019},
  track_display_name = {Technical Briefs},
  icon = {/assets/img/neural-material-vis.png},
  genesis = {/flexible_nr_genesis}
}