“Multi-scale shape manipulations in photographs” by Orzan, Barla, Bousseau and Thollot
Conference:
Type(s):
Title:
- Multi-scale shape manipulations in photographs
Presenter(s)/Author(s):
Abstract:
Visual content is often better communicated by simplified or exaggerated images than by the “real world like” photographs. We propose a tool for manipulating the shape of image objects and their level of detail, while still preserving the original image “look”. Recent research has built on image information to create very stylized renderings [Kang et al. 2005]. While this approach greatly helps the user to create new representations, it still relies a lot on artistic skills and the resulting image only bears some resemblance with the original input. In contrast, we are interested in keeping most of the information present in the image, while still allowing a non artist to intuitively modify the shape of depicted objects. Our approach is thus to rely on details and color variations already present in the input, hence avoiding tedious editing operations.
References:
1. Elder, J. H., and Goldberg, R. M. 2001. Image editing in the contour domain. IEEE Transactions on Pattern Analysis and Machine Intelligence 23, 3, 291–296.
2. Kang, H. W., Chui, C. K., and Chakraborty, U. K. 2005. Interactive sketch generation. The Visual Computer (Proceedings of Pacific Graphics 2005) 21, 8, 821–830.
Additional Images:
