Revision history [back]

Hello everyone, I have been given a task to train general purpose face detector using haar cascade. My current approach, is to detect faces in the CelebA dataset using INTEL FRONTAL FACE CASCADES. My positive set contain 6000 face crops. My negative set contains 10,000 images of mostly aircrafts. QUESTIONS: 1) I wanted to know atleast how many images should be in each set to get good results? 2) after training with current setup i get "leaf hitrate achieved, branch training is terminated" is this normal. 3) Is CelebA dataset even suitable for this purpose? 4) does using negatives of sametype(aircrafts) affect accuracy? Does resolution of negatives matter?

Note : Face Crops are all square and of same size. Resolution of negatives is around 300x180.

Training parameters: opencv_traincascade -info info.lst -vec positives.vec w-24 -h 24 -numPos 5200 -numNeg 10000 -numStages 10 -minHitRate 0.998 -minFalseAlarmRate 0.25

Hello everyone, I have been given a task to train general purpose face detector using haar cascade. My current approach, is to detect faces in the CelebA dataset using INTEL FRONTAL FACE CASCADES. My positive set contain 6000 face crops. My negative set contains 10,000 images of mostly aircrafts. QUESTIONS: 1) QUESTIONS:

1. I wanted to know atleast how many images should be in each set to get good results? 2) results?
2. after training with current setup i get "leaf hitrate achieved, branch training is terminated" is this normal. 3) normal.
3. Is CelebA dataset even suitable for this purpose? 4) purpose?
4. does using negatives of sametype(aircrafts) affect accuracy? Does resolution of negatives matter?

Note : Face Crops are all square and of same size. Resolution of negatives is around 300x180.

Training parameters: opencv_traincascade -info info.lst -vec positives.vec w-24 -h 24 -numPos 5200 -numNeg 10000 -numStages 10 -minHitRate 0.998 -minFalseAlarmRate 0.25

Hello everyone, I have been given a task to train general purpose face detector using haar cascade. My current approach, is to detect faces in the CelebA dataset using INTEL FRONTAL FACE CASCADES. My positive set contain 6000 face crops. My negative set contains 10,000 images of mostly aircrafts. QUESTIONS:

1. I wanted to know atleast how many images should be in each set to get good results?
2. after training with current setup i get "leaf hitrate achieved, branch training is terminated" is this normal.
3. Is CelebA dataset even suitable for this purpose?
4. does using negatives of sametype(aircrafts) affect accuracy? accuracy?
5. Does resolution of negatives matter?

Note : Face Crops are all square and of same size. Resolution of negatives is around 300x180.

Training parameters: opencv_traincascade -info info.lst -vec positives.vec w-24 -h 24 -numPos 5200 -numNeg 10000 -numStages 10 -minHitRate 0.998 -minFalseAlarmRate 0.25