Translated Title
Stereoscopic Image Color Correction Based on Matching Features Fusion
Translated Abstract
Stereoscopic image color correction aims to eliminate the color difference between the left and right views of the stereoscop-ic image.Existing stereo image color correction methods have problems in balancing correction effect and time efficiency.To solve this problem,we proposed a stereoscopic image color correction method based on matching features fusion.First,the parallax attention col-or correction network is used to obtain the initial correction result.Then,the initial correction result and the reference image are input into the optical flow-based image matching network to obtain the matching target image.Finally,fusing the initial correction result,the matching target image,the reference image,and the target image to obtain the final correction result through the image fusion network.Experiments show that the method proposed in this paper has advanced performance and can achieve high-quality stereoscopic image color correction while maintaining high time efficiency.
Translated Keyword
color correction
convolutional neural network
image fusion
stereoscopic image
Access Number
WF:perioarticalxxwxjsjxt202401034
