CV_open's profile - activity

For the EAST model provided with openCV, which dataset was used in the training process?

EAST text detector For the EAST model provided with openCV, which dataset was used in the training process?

To create the haar cascade classifier, if I train it with images of only 1 object compared to many objects of the same class, will it be able to detect only that specific object that I trained it with or providing that there is not much variance within the class, will the resulting cascade classifier actually detect other objects within the class?

I am using the traincascade.exe from OpenCV 3.1 and trying to speed up the processing time.

I set the precalcValBufSize and the precalcIdxBufSize both to 512 and then both to 2048. However the run time to create the classifier did not change from when no value was specified for these values (using default value of 1024), is there a problem with precalcValBufSize and the precalcIdxBufSize in OpenCV 3.1?

I have trained a haar cascade using traincascade and put -acceptanceRatioBreakValue 0.0001. The training then stopped saying that the required acceptance ratio has been reached. So at stage 9,

NEG count : acceptanceRatio 10000 : 5.86107e-05 The required acceptanceRatio for the model has been reached to avoid overfitting of trainingdata. Branch training terminated.

My question is, if it is correct that the acceptance ratio is calculated for the previous stage ie stage 8 in this case, : - should I run the traincascade command again so that stage 8 is removed from the xml file?