CG2Real: Improving the Realism of Computer Generated Images Using a Large Collection of Photographs

118 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Learning from Simulated and Unsupervised Images through Adversarial Training (2016, background): “[16] transfer the style from a set of real images to the synthetic image by co-segmenting and then identifying similar regions.”
Semi-Parametric Image Synthesis (2018, background): “[13] described a related system for post-processing computer-generated images.”
Deep Image Harmonization (2017, methodology): “On the other hand, a data-driven method [10] is developed to improve the realism of computer-generated images by retrieving a set of real images with similar global layouts for transferring appearances.”
Data-driven visual similarity for cross-domain image matching (2011, background): “2009], and even making CG imagery more realistic [Johnson et al. 2010], to give but a few examples.”
Visual attribute transfer through deep image analogy (2017, result): “We further compare our methods with the CG2Real [Johnson et al. 2011] on their provided examples, shown in Figure 18.”
Enhancing Photorealism Enhancement (2021, background): “improve the realismof rendered images by transferring nearest neighbor patches from similarly structured photographs [45].”
Learning a Discriminative Model for the Perception of Realism in Composite Images (2015, background): “These include transition of colors from the foreground to the background [1,20], color inconsistencies [15,23,24,33], texture inconsistencies [4, 11], and suppressing “bleeding” artifacts [31].”