Capturing Light with Robots: A Novel Workflow for Reproducing Realistic Lens Flares

Filmakademie Baden-Württemberg, Animationsinstitut
CVMP 2024

Abstract

Lens flares are a common optical artifact in photography and filmmaking caused by reflections and scattering within a camera system. While they can degrade image quality, lens flares are sometimes harnessed artistically to enhance visual appeal. This reasearch captures lens flares using a motion control setup. It harnesses image based and machine learning interpolation techniques to represent the flare artifacsts faithfully in compositing.

Nuke lens flare picker

Interpolation comparison

Custom CNN implemented in Pytorch/ Nuke Cattery

Explanation Video

Poster

Dataset

BibTeX


@inproceedings{10.1145/3681758.3697995,
author = {Maurer, Vincent},
title = {Capturing Light with Robots: A Novel Workflow for Reproducing Realistic Lens Flares},
year = {2024},
isbn = {9798400711404},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3681758.3697995},
doi = {10.1145/3681758.3697995},
abstract = {Lens flares are a common optical artifact in photography and filmmaking caused by reflections and scattering of light within a camera system. The task of creating lens flares is usually solved with 2D approaches or simulation. This research compares two alternative methods of reproducing lens flares for a production-ready compositing workflow: traditional image processing and machine learning techniques. To create the dataset, a novel approach to capturing lens flares in a grid-like manner is explored. By systematically varying the position of a light source with a motion control system, flare patterns for a diverse set of lenses are captured.},
booktitle = {SIGGRAPH Asia 2024 Technical Communications},
articleno = {20},
numpages = {4},
keywords = {Lens Flares, Visual Effects, Compositing, Motion Control, Machine Learning},
location = {
},
series = {SA '24}
}