没有合适的资源?快使用搜索试试~ 我知道了~
基于Lowe在04年的论文做的总结,其中还有若干不清楚的地方稍后会补充
资源详情
资源评论
资源推荐
SIFT review
by fei sun
School of Mathematics, Peking University
What is SIFT?
SIFT = Scale Invariant Feature Transform
invariant to image scaling, translation, and rotation,
partially invariant to illumination changes and affine or 3D projection.
Primarly invented and developed by David G. Lowe in 1999 and 2004.
Reference:
David G. Lowe, "Distinctive image features from scale-invariant keypoints," International
Journal of Computer Vision, 60, 2 (2004), pp. 91-110.
David G. Lowe, "Object recognition from local scale-invariant features," International Con-
ference on Computer Vision, Corfu, Greece (September 1999), pp. 1150-1157.
David G. Lowe, "Local feature view clustering for 3D object recognition," IEEE Conference
on Computer Vision and Pattern Recognition,Kauai, Hawaii (December 2001), pp. 682-688.
For image matching and recognition, SIFT features are first extracted from a set of reference
images and stored in a database. A new image is matched by individually comparing each
feature from the new image to this previous database and finding candidate matching features
based on Euclidean distance of their feature vectors.
SIFT review
1 Overall of SIFT
1. Scale-space extrema detection
identify potential interest points that are invariant to scale and orientation
2. Keypoint localization
selected by stability
3. Orientation assignment
One or more orientations are assigned to each keypoint location based on local image
gradient directions.
4. Keypoint descriptor
each feature is transformed relative to the assigned orientation, scale, and location
SIFT review
2 Scale-space extrema detection
Definition 1. Scale space
L(x; y; σ) = G(x; y; σ ) ∗I(x; y)
where I(x; y) is the input image, G(x; y; σ) is the Gaussian function
G(x; y; σ) =
1
2πσ
2
e
¡(x
2
+y
2
)/2σ
2
WHY scale-space:
Detecting locations that are invariant to scale change of the image can be accomplished by
searching for stable features across all possible scales, using a continuous function of scale
known as scale space (Witkin, 1983).
Gaussian kernel (and its derivatives): the only possible smoothing kernels for scale space
analysis.
ref: Koenderink, J.J. 1984. The structure of images. Biological Cybernetics, 50:363–396.
剩余15页未读,继续阅读
misssharp
- 粉丝: 0
- 资源: 1
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功
评论0