Recurrent Color Constancy Supplementary Material
نویسندگان
چکیده
Groundtruth Bias – We render all images in the linear SFU Gray Ball Dataset with respect to the linear groundtruth provided by the color constancy benchmarking website [1]. Among all rendered images, we find a number of these not in intrinsic color and not consistent in illumination with the preceding frames (Figure 1), which indicates the corresponding “groundtruth” is biased. The biased “groundtruth” makes our algorithm not accurately estimate the illumination, which is reflected in maximum angular error we reported in the main paper. More Results – To better understand the superiority of our RCC-Net in temporal color constancy task, we collect some statistics (angular error and angle change) of illumination estimation for 20 five-frame sequences from the Gray Ball Dataset, of which 10 sequences (Figure 2) are with significant illumination change and the remainder (Figure 3) are with minor illumination change. It is easily observable that regardless of varying-illumination or constant-illumination conditions, our RCC-Net usually notably outperforms others. In addition, 6 samples of color correction by temporal methods are shown in Figure 4. Generally, corrected images using estimation given by the RCC-Net seems more similiar to the groundtruth image, compared to those images corrected using predicted illuminations from other temporal methods. Note that: for the sake of convenience, all experiments and results here are correspond to the linear version of the Gray Ball Dataset.
منابع مشابه
FC: Fully Convolutional Color Constancy with Confidence-weighted Pooling Supplementary Material
متن کامل
The effects of surface gloss and roughness on color constancy for real 3-D objects.
Color constancy denotes the phenomenon that the appearance of an object remains fairly stable under changes in illumination and background color. Most of what we know about color constancy comes from experiments using flat, matte surfaces placed on a single plane under diffuse illumination simulated on a computer monitor. Here we investigate whether material properties (glossiness and roughness...
متن کاملColor constancy: phenomenal or projective?
Naive observers viewed a sequence of colored Mondrian patterns, simulated on a color monitor. Each pattern was presented twice in succession, first under one daylight illuminant with a correlated color temperature of either 16,000 or 4000 K and then under the other, to test for color constancy. The observers compared the central square of the pattern across illuminants, either rating it for sam...
متن کاملEdge Classification for Color Constancy
The goal of color constancy is to measure image colors despite differences in the color of the light source. Traditionally, the computational method of obtaining this ability is by using pixel values only. Recently, methods using edges instead of pixel values have been proposed. However, different edge types exist, such as material, shadow and specular edges. Therefore, in this paper, the main ...
متن کاملThe Ratio Method for Multi-view Color Constancy
Color constancy is the ability to infer stable material colors despite changes in lighting, and it is typically addressed computationally using a single image as input. In many recognition and retrieval applications, we have access to image sets that contain multiple views of the same object in different environments; we show in this technical report and a related publication [8], that correspo...
متن کامل