CS231A Lecture 9：Fitting and Matching

lecture 9 也没什么内容，不过还是好好记一下，霍夫变换之前不太了解。。。

Reading:

[HZ] Chapter: 4 “Estimation – 2D projective transformation”

[HZ] Chapter: 11 “Computation of the fundamental matrix F”

[FP] Chapter:10 “Grouping and model fitting”

Problem formulation

Fitting (拟合)

Least square methods

最小二乘法

CONCLUSION: Least square is not robust w.r.t. outliers

RANSAC

见：使用RANSAC算法拟合点云中的平面

这里仅补充几个需要关注的地方。

RANSAC的两个基本假设

How many samples?

Conclusions

Hough transforms

P.V.C. Hough, Machine Analysis of Bubble Chamber Pictures, Proc. Int. Conf. High Energy Accelerators and Instrumentation, 1959

Conclusions

Generalized Hough transform

D. Ballard, Generalizing the Hough Transform to Detect Arbitrary Shapes, Pattern Recognition 13(2), 1981

思考：

课件里面霍夫变换讲的并不清楚，这里补充一些内容：

• 霍夫变换——神奇的特征提取算法
• 史上最全霍夫变换详解（含源码）
• 霍夫圆检测原理

Multi-model fitting

Fitting multiple models

Incremental line fitting