- 最后登录
- 2017-9-18
- 注册时间
- 2011-1-12
- 阅读权限
- 90
- 积分
- 12276
  
- 纳金币
- 5568
- 精华
- 0
|
1 Introduction
Extraction and re-rendering of real materials give large contributions
to various image-based applications. As one of the key properties
of modeling the appearance of an object, materials mainly
focus on the effects caused by light transportation. Therefore, understanding
the characteristics of a complex material from a single
photograph and transferring it to an object in another image becomes
a very challenging problem.
Khan et al. [2006] present an image-based material editing method
which can produce translucent or transparent materials in place of
the original diffuse objects or re-render the objects with arbitrary
BRDFs in a single high dynamic range photograph. However, the
experiential models in their system sometimes is not convenient to
create very realistic images, using real materials as input should
be considered. Moreover, this approach employs a re-texturing
method instead of simulating the real mechanism of light transport
in translucent material, that will also affect the synthesis quality.
In this paper, we present a novel framework to transfer real translucent
materials such as fruits and flowers between single images. We
define a group of information which can model the attributes during
the extraction and transfer process. Once we extract those information
from both the source and target images, we can easily produce
a realistic photograph of an object with target-like materials and
suitable shading effects in the environment of sources. |
|