Rendering Synthetic Objects into Legacy Photographs
Karsch, K.; Hedau, V.; Forsyth, D.; Hoiem, D.
ACM Transactions on Graphics (Proceedings of SIGGRAPH Asia), 30(6), 2011. doi: 10.1145/2024156.2024191
The RSOLP Annotator is a web interface that provides a basic implementation of the method detailed in the Rendering Synthetic Objects into Legacy Photographs (RSOLP) paper. This software allows you to upload a picture and semi-automatically create a rough 3D model of scene. The method is best suited for photographs of rooms and other man-made scenes, but may work in other cases as well. Through simple interactions, the interface allows an image to be annotated quickly and then sent to our server for processing. Once the 3D scene has been computed, you will receive an email containing a link to download your reconstructed scene. For references and examples, please refer to the Help tab within the Annotator's interface. This software is not for commercial use.
If you use this software in your research, please cite the following paper:
If you find the RSOLP Annotator useful, you might also want to try the RSOLP Compositor for producing rendered images and compositing final results.
For software updates and other relevant information pertaining to RSOLP, join the rsolp mailing group by submitting your email address below.