Christosh's profile - activity

How can I convert a vector of Mat files to a Mat file? I have got vectror<mat> with size N. Exery mat file is of size 1xM. I want to create a new Mat file of size 1xN*M? How can I do so in opencv?

How can I convert a vector of Mat files to a Mat file? I have got vectror<mat> with size N. Exery mat file is of size 1xM. I want to create a new Mat file of size 1xN*M? How can I do so in opencv?

Is it necessary the use of perl scripting?

I want to use opencv and haartraining in order to train a new haarcascade with new samples. I have created positives.dat and negative.dat which contains list with positive and negative images. Next step to train haarcascade using opencv_haartraining.exe. However I am confused with documentation here. Any idea how can I use haartraining.exe?

I want to extract foreground and background of images. I am interesting in foreground with humans. So first step is face detections algorithm. The next step the use of grab_cut algorithm in order to extract foreground. How is it possible to use the information from face detection in order to automatically extract the foreground of image. I am familiar with grab_cut python implementation where you select mannually the region where is your foreground. How can I do this mannually using face detector??

I am trying to detect body from images using haar cascades. I am using python code in order to do so, my code is the following:

import numpy as np
import sys
import cv2

body_cascade = cv2.CascadeClassifier('cascades/haarcascade_upperbody.xml')
img = cv2.imread('body/body4.jpg')
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

faces = body_cascade .detectMultiScale(gray, 1.1, 8)

for (x,y,w,h) in faces:
   cv2.rectangle(img,(x,y),(x+w,y+h),(0,255,0),2)
   roi_gray = gray[y:y+h, x:x+w]
   roi_color = img[y:y+h, x:x+w]

I ve tested several images, the three cascades for body several scalefactors. However, it seems not finding the body in any image. Am I doing something wrong in the parameters of detection? When I tried frontface cascades everything worked fine.

I I am trying to use createEigenFaceRecognizer() in Labeled Faces in the Wild Home database. I have trained a model using suggested code. When I test the model with the trained samples I am receiving a 100% success, however, with unknown samples I am receiving a success rate close to 20%. Is there any standar possible reason for this overtrain?? I am using 40 different persons with 5-8 images per person(for training) and one for test purposes. I am using the function createEigenFaceRecognizer(0) in order to use all the eigenvalues.