Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

On Finding Gray Pixels

About

We propose a novel grayness index for finding gray pixels and demonstrate its effectiveness and efficiency in illumination estimation. The grayness index, GI in short, is derived using the Dichromatic Reflection Model and is learning-free. GI allows to estimate one or multiple illumination sources in color-biased images. On standard single-illumination and multiple-illumination estimation benchmarks, GI outperforms state-of-the-art statistical methods and many recent deep methods. GI is simple and fast, written in a few dozen lines of code, processing a 1080p image in ~0.4 seconds with a non-optimized Matlab code.

Yanlin Qian, Joni-Kristian K\"am\"ar\"ainen, Jarno Nikkanen, Jiri Matas• 2019

Related benchmarks

TaskDatasetResultRank
Color ConstancyNCC (in-dataset)
Median Error3.13
29
Color ConstancyLEVI (in-dataset)
Median Error3.1
24
Color ConstancyGehler-Shi
Median Error1.87
22
Illuminant EstimationGehler-Shi (test)
Mean Error3.07
21
Illuminant EstimationNUS-8 (test)
Mean Error2.91
21
Illuminant Estimation (Recovery)ColorChecker REC (test)
Median Error1.91
20
Illuminant EstimationIntel_TAU (full dataset)
Recovery Median Error2.46
20
Color ConstancyETH3D RAW
Mean Error3.26
19
Color ConstancyDepthAWB
Mean Error3.91
19
Color ConstancyNYU-v2 & Diode
Mean Angular Error4.28
19
Showing 10 of 16 rows

Other info

Follow for update