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

Authors:
Micah K. Johnson;Kevin Dale;Shai Avidan;Hanspeter Pfister;William T. Freeman;Wojciech Matusik
Affiliations:
Massachusetts Institute of Technology, Cambridge;Harvard University, Cambridge;Tel Aviv University, Israel, and Adobe Systems;Harvard University, Cambridge;Massachusetts Institute of Technology, Cambridge;Massachusetts Institute of Technology, Cambridge
Venue:
IEEE Transactions on Visualization and Computer Graphics
Year:
2011

Citing 0
Cited 8

Data-driven visual similarity for cross-domain image matching

Proceedings of the 2011 SIGGRAPH Asia Conference
Data-Driven Object Manipulation in Images

Computer Graphics Forum
3D Material Style Transfer

Computer Graphics Forum
IMShare: instantly sharing your mobile landmark images by search-based reconstruction

Proceedings of the 20th ACM international conference on Multimedia
Technical Section: Automatic color realism enhancement for computer generated images

Computers and Graphics
Example-based video color grading

ACM Transactions on Graphics (TOG) - SIGGRAPH 2013 Conference Proceedings
Accurate and efficient cross-domain visual matching leveraging multiple feature representations

The Visual Computer: International Journal of Computer Graphics
PatchNet: a patch-based image representation for interactive library-driven image editing

ACM Transactions on Graphics (TOG)

Quantified Score

Hi-index	0.00

Visualization

Abstract

Computer-generated (CG) images have achieved high levels of realism. This realism, however, comes at the cost of long and expensive manual modeling, and often humans can still distinguish between CG and real images. We introduce a new data-driven approach for rendering realistic imagery that uses a large collection of photographs gathered from online repositories. Given a CG image, we retrieve a small number of real images with similar global structure. We identify corresponding regions between the CG and real images using a mean-shift cosegmentation algorithm. The user can then automatically transfer color, tone, and texture from matching regions to the CG image. Our system only uses image processing operations and does not require a 3D model of the scene, making it fast and easy to integrate into digital content creation workflows. Results of a user study show that our hybrid images appear more realistic than the originals.