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

Learning a Discriminative Model for the Perception of Realism in Composite Images

About

What makes an image appear realistic? In this work, we are answering this question from a data-driven perspective by learning the perception of visual realism directly from large amounts of data. In particular, we train a Convolutional Neural Network (CNN) model that distinguishes natural photographs from automatically generated composite images. The model learns to predict visual realism of a scene in terms of color, lighting and texture compatibility, without any human annotations pertaining to it. Our model outperforms previous works that rely on hand-crafted heuristics, for the task of classifying realistic vs. unrealistic photos. Furthermore, we apply our learned model to compute optimal parameters of a compositing method, to maximize the visual realism score predicted by our CNN model. We demonstrate its advantage against existing methods via a human perception study.

Jun-Yan Zhu, Philipp Kr\"ahenb\"uhl, Eli Shechtman, Alexei A. Efros• 2015

Related benchmarks

TaskDatasetResultRank
Image HarmonizationiHarmony4 HFlickr
MSE315.4
58
Image HarmonizationiHarmony4 (all)
MSE204.8
53
Image HarmonizationiHarmony4 Hday2night
MSE136.7
51
Image HarmonizationiHarmony4 HAdobe5k
MSE414.3
43
Image HarmonizationS-Adobe5K (test)
MSE41.29
25
Image HarmonizationiHarmony4 HCOCO
MSE79.82
20
Image Harmonization99 real composite images (test)
B-T Score0.337
12
Image HarmonizationiHarmony4 0%-5% foreground ratio
MSE33.3
12
Image HarmonizationiHarmony4 5%-15% foreground ratio
MSE145.1
12
Image HarmonizationiHarmony4 15%-100% foreground ratio
MSE682.7
12
Showing 10 of 18 rows

Other info

Follow for update