DavidBankom1's profile - activity

overview network karma followed questions activity

2017-04-12 10:07:30 -0600	asked a question	Order of sampling negative examples Lets say I have many images for negative sampling listed in the background file. But the resolution of these images is so high (720 by 1280) compared to the size of sliding window (15 by 21) that even one image quantitatively contains enough negative examples for training the whole cascade. Does this mean traincascade will sample all negative examples from the picture which is listed first in the background file?
2017-04-07 19:04:04 -0600	commented answer	Meaning of stages in traincascade @StevenPuttemans thanks for your comment. However it's not quite an answer to my question. What's the way of combining stumps inside one stage? Are they applied sequentially? If yes then I don't see any meaning in staging the process, because these weak classifiers themselves are working as little stages.
2017-04-07 13:49:28 -0600	asked a question	Meaning of stages in traincascade I'm wondering what the special meaning of stages is. As far as I understood every stage might consist of arbitrary number of classifiers depending on specified parameters of launch like `minHitRate`. Thus the exact depth of the tree can't be specified directly by the user and inferred during the training. Why didn't developers just call weak classifiers as 'stages' ? I know that stages are applied sequentially weeding out negatives but single weak classifiers could be applied in a same fashion.
2017-04-04 05:51:00 -0600	received badge	● Editor (source)
2017-04-04 03:51:32 -0600	commented question	training classifier freezes @berak no problem, thanks for clarifying this
2017-04-03 18:39:01 -0600	asked a question	training classifier freezes I posted similar question a few hours ago. After that my account was blocked without any explanations. If you are administrator and you're reading this, I hope you will at least give me a clue what's wrong with my question after deleting it second time. So here it is: I'm trying to train my own cascade for face detection in opencv using preexisting `.vec` file with humans faces provided with opencv package. I sampled 60 negative examples described in `bg.txt` (however essentially there're only 3 different pictures, I just copied them many times). It's not much but I just want to see that cascade is created and later expand the training set. So I run the following command opencv_traincascade -data data -vec faces.vec -bg bg.txt -numPos 100 -numNeg 50 -numStages 1 -w 24 -h 24 -featureType LBP Then the following information gets printed in the terminal PARAMETERS: cascadeDirName: data vecFileName: faces.vec bgFileName: bg.txt numPos: 100 numNeg: 50 numStages: 2 precalcValBufSize[Mb] : 1024 precalcIdxBufSize[Mb] : 1024 acceptanceRatioBreakValue : -1 stageType: BOOST featureType: LBP sampleWidth: 24 sampleHeight: 24 boostType: GAB minHitRate: 0.995 maxFalseAlarmRate: 0.5 weightTrimRate: 0.95 maxDepth: 1 maxWeakCount: 100 Number of unique features given windowSize [24,24] : 8464 ===== TRAINING 0-stage ===== POS count : consumed 100 : 100 NEG count : acceptanceRatio 50 : 1 After that nothing happens. No errors, it just freezes. It uses CPU intensively though. Any thoughts on this will be much appreciated. Thanks