FLIP: A Difference Evaluator for Alternating Images

Forskningsoutput: TidskriftsbidragArtikel i vetenskaplig tidskrift

Standard

FLIP: A Difference Evaluator for Alternating Images. / Andersson, Pontus; Akenine-Möller, Tomas; Nilsson, Jim; Åström, Kalle; Oskarsson, Magnus; Fairchild, Mark.

I: Proceedings of the ACM in Computer Graphics and Interactive Techniques, Vol. 3, Nr. 2, 15, 08.2020.

Forskningsoutput: TidskriftsbidragArtikel i vetenskaplig tidskrift

Harvard

APA

CBE

MLA

Vancouver

Author

RIS

TY - JOUR

T1 - FLIP: A Difference Evaluator for Alternating Images

AU - Andersson, Pontus

AU - Akenine-Möller, Tomas

AU - Nilsson, Jim

AU - Åström, Kalle

AU - Oskarsson, Magnus

AU - Fairchild, Mark

PY - 2020/8

Y1 - 2020/8

N2 - Image quality measures are becoming increasingly important in the field of computer graphics. For example, there is currently a major focus on generating photorealistic images in real time by combining path tracing with denoising, for which such quality assessment is integral. We present FLIP, which is a difference evaluator with a particular focus on the differences between rendered images and corresponding ground truths. Our algorithm produces a map that approximates the difference perceived by humans when alternating between two images. FLIP is a combination of modified existing building blocks, and the net result is surprisingly powerful. We have compared our work against a wide range of existing image difference algorithms and we have visually inspected over a thousand image pairs that were either retrieved from image databases or generated in-house. We also present results of a user study which indicate that our method performs substantially better, on average, than the other algorithms. To facilitate the use of FLIP, we provide source code in C++, MATLAB, NumPy/SciPy, and PyTorch.

AB - Image quality measures are becoming increasingly important in the field of computer graphics. For example, there is currently a major focus on generating photorealistic images in real time by combining path tracing with denoising, for which such quality assessment is integral. We present FLIP, which is a difference evaluator with a particular focus on the differences between rendered images and corresponding ground truths. Our algorithm produces a map that approximates the difference perceived by humans when alternating between two images. FLIP is a combination of modified existing building blocks, and the net result is surprisingly powerful. We have compared our work against a wide range of existing image difference algorithms and we have visually inspected over a thousand image pairs that were either retrieved from image databases or generated in-house. We also present results of a user study which indicate that our method performs substantially better, on average, than the other algorithms. To facilitate the use of FLIP, we provide source code in C++, MATLAB, NumPy/SciPy, and PyTorch.

UR - https://research.nvidia.com/publication/2020-07_FLIP

M3 - Article

VL - 3

JO - Proceedings of the ACM in Computer Graphics and Interactive Techniques

JF - Proceedings of the ACM in Computer Graphics and Interactive Techniques

SN - 2577-6193

IS - 2

M1 - 15

ER -