Training Process of Adaboost
In recent days, I have been done to modify the training process of adaboost of OpenCV with D2MCE. D2MCE is a distributed mobile computing environment which prov …
difference
In recent days, I have been done to modify the training process of adaboost of OpenCV with D2MCE. D2MCE is a distributed mobile computing environment which prov …
出處:SiPS 2008. IEEE Workshop on Signal Processing Systems, 2008. 論文名稱:Parallelization of AdaBoost algorithm on multi-core processors 作者:Yen-Kuang Chen Wenlong …
Tutorial: OpenCV haartraining (Rapid Object Detection With A Cascade of Boosted Classifiers Based on Haar-like Features) http://note.sonots.com/SciSoftware/haar …
假設有M個臉部影像G1,G2,G3,…,GM ,每個影像的大小為N*N,運作程序如下: 1. 將一個臉部的二維影像資料轉成一維陣列的型式 2. 計算平均臉部影像,式子如下: Ψ = (1/M) ∑Mi=1 Gi 3. 計算每個「臉部影像」與「平均臉部影像」的差異: Φi = Γi – Ψ 4. …
K=1, 拿最接近的點作為自己的標記類型(將自己標為此類型的點) K=3, 拿三個最接近的點,看三個裡面哪個點最多,就標記為哪種點
以下是取材自wiki的資料。 http://en.wikipedia.org/wiki/Cross-validation K-fold cross-validation In K-fold cross-validation, the original sample is partitioned into K subsa …
起始點是從「如何用一個點來表示空間中的兩個點?」可以用取平均值的方式。 PCA是用「Least Squared Error」的角度來求解。 如何找出一個向量滿足squared-error criterion function? 式子導一導就變成covariance matrix了,這時候為了求裡面的一個參數,所以導入L …