- 本文主要介绍了
Selective Search
算法,该算法被广泛应用于物体检测算法中。
-
整体流程
Algorithm 1: Hierarchical Grouping algorithm Input: (color)image Output: Set of object location hypotheses Obtain initial tegions R = {r1,...,rn} using[1] Initialise similarity set S = 0 foreach Neighbouring region pair(ri, rj) do Calculate similarity s(ri,rj) S = S∪s(ri,rj) while S != ∅ do Get highest similarity s(ri,rj) = max(S) Merge corresponding regions rt = ri ∪ rj Remove similarities regarding ri:S = S\s(ri,r*) Remove similarities regarding rj:S = S\s(r*,rj) Calculate similarity set St between rt and its neighbours S = S ∪ St R = R ∪ rt Extract object location boxes L from all regions in R
-
首先
input:
一张W*H*3
的图片 -
output:
一组由边界组成的集合 -
获取初始化图像分割使用
Efficient Graph-Based Image Segmentation
-
S表示的是所有区域之间的相似度,不断合并其中相似度最高的区域
- colour(ri,rj)
- texture(ri,rj)
- 使用
SIFT
算法进行梯度提取总和
- 使用
- size(ri,rj)
- fill(ri,rj)
- 计算四个相似度之和,按照整体流程描述的合并思路进行合并
s_colour(r_i,r_j) = \sum^{n}_{k=1}\min(c_i^k,c_j^k)
C_t = \frac{size(r_i)*C_i+size(r_j)*C_j}{size(r_i)+size(r_j)}
s_{size}(r_i,r_j) = 1 - \frac{size(r_i)+size(r_j)}{size(im)}
fill(r_i,r_j) = 1-\frac{size(BB_ij)-size(r_i)-size(r_i)}{size(im)}