Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
PointCloud
Flann
提交
56dc18db
F
Flann
项目概览
PointCloud
/
Flann
通知
4
Star
1
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
0
列表
看板
标记
里程碑
合并请求
0
DevOps
流水线
流水线任务
计划
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
F
Flann
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
0
Issue
0
列表
看板
标记
里程碑
合并请求
0
合并请求
0
Pages
DevOps
DevOps
流水线
流水线任务
计划
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
流水线任务
提交
Issue看板
体验新版 GitCode,发现更多精彩内容 >>
提交
56dc18db
编写于
4月 17, 2008
作者:
M
Marius Muja
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Readme file
上级
fa129857
变更
2
显示空白变更内容
内联
并排
Showing
2 changed file
with
70 addition
and
0 deletion
+70
-0
README
README
+7
-0
src/algo/KMeansTree.d
src/algo/KMeansTree.d
+63
-0
未找到文件。
README
0 → 100644
浏览文件 @
56dc18db
FLANN - Fast Library for Approximate Nearest neighbors
======================================================
This is a library for fast approximate nearest neighbor matching.
See the doc/manual.pdf file for information on how to use the library.
src/algo/KMeansTree.d
浏览文件 @
56dc18db
...
...
@@ -353,6 +353,69 @@ class KMeansTree(T) : NNIndex
return
centers
[
0.
.
index
];
}
/+
vector<Point> chooseSmartCenters(const vector<Point> &data, int numCenters, int numLocalTries) {
ASSERT( numCenters > 0 && numCenters <= (int)data.size() );
int i;
Scalar currentPot = 0;
vector<Point> centers;
vector<Scalar> closestDistSq;
// Choose one random center and set the closestDistSq values
int index = (int)(getRandomScalar() * data.size());
centers.push_back(data[index]);
for (i = 0; i < (int)data.size(); i++) {
closestDistSq.push_back( distSq(data[i], data[index]) );
currentPot += closestDistSq[i];
}
// Choose each center
for (int centerCount = 1; centerCount < numCenters; centerCount++) {
// Repeat several trials
Scalar bestNewPot = -1;
int bestNewIndex;
for (int localTrial = 0; localTrial < numLocalTries; localTrial++) {
// Choose our center - have to be slightly careful to return a valid answer even accounting
// for possible rounding errors
Scalar randVal = getRandomScalar() * currentPot;
for (index = 0; index < (int)data.size()-1; index++) {
if (randVal <= closestDistSq[index])
break;
else
randVal -= closestDistSq[index];
}
// Compute the new potential
Scalar newPot = 0;
for (i = 0; i < (int)data.size(); i++)
newPot += min( distSq(data[i], data[index]), closestDistSq[i] );
// Store the best result
if (bestNewPot < 0 || newPot < bestNewPot) {
bestNewPot = newPot;
bestNewIndex = index;
}
}
// Add the appropriate center
centers.push_back(data[bestNewIndex]);
currentPot = bestNewPot;
for (i = 0; i < (int)data.size(); i++)
closestDistSq[i] = min( distSq(data[i], data[bestNewIndex]), closestDistSq[i] );
}
return centers;
}
+/
/**
* Builds the index
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录