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

DoveNet: Deep Image Harmonization via Domain Verification

About

Image composition is an important operation in image processing, but the inconsistency between foreground and background significantly degrades the quality of composite image. Image harmonization, aiming to make the foreground compatible with the background, is a promising yet challenging task. However, the lack of high-quality publicly available dataset for image harmonization greatly hinders the development of image harmonization techniques. In this work, we contribute an image harmonization dataset iHarmony4 by generating synthesized composite images based on COCO (resp., Adobe5k, Flickr, day2night) dataset, leading to our HCOCO (resp., HAdobe5k, HFlickr, Hday2night) sub-dataset. Moreover, we propose a new deep image harmonization method DoveNet using a novel domain verification discriminator, with the insight that the foreground needs to be translated to the same domain as background. Extensive experiments on our constructed dataset demonstrate the effectiveness of our proposed method. Our dataset and code are available at https://github.com/bcmi/Image_Harmonization_Datasets.

Wenyan Cong, Jianfu Zhang, Li Niu, Liu Liu, Zhixin Ling, Weiyuan Li, Liqing Zhang• 2019

Related benchmarks

TaskDatasetResultRank
Image HarmonizationiHarmony4 HFlickr
MSE133.1
58
Image HarmonizationiHarmony4 (all)
MSE52.33
53
Image HarmonizationiHarmony4 Hday2night
MSE51.95
51
Image HarmonizationiHarmony4 HAdobe5k
MSE52.32
43
Image HarmonizationiHarmony4 HCOCO
MSE36.72
38
Image HarmonizationHAdobe5k iHarmony4 (test)
MSE51
37
Image HarmonizationiHarmony4
MSE52.33
27
Image HarmonizationS-Adobe5K (test)
MSE52.32
25
Image HarmonizationiHarmony4 HCOCO
MSE36.72
20
Image HarmonizationDIH99 (test)
Average Processing Time (s)0.05
17
Showing 10 of 46 rows

Other info

Code

Follow for update