Revision history [back]

it's amazing, what kind of things folks are trying to train a cascade on ..

i'm not trying to discourage you entirely, trying this, but your attempt suffers from:

• pose problem. you'd need to train 2 classifiers, like cow_frontal, and cow_profile (and flip images for the latter)
• not enough positives. you'd need like ~100 times more
• -w=24 -h=24, that's probably too small, and the wrong shape(distortion). w=64, h=32 might work much better

also note, there's a pretrained cow detector in opencv_contrib/dpm, also there are pretrained dnn's that can detect cows

it's amazing, what kind of things folks are trying to train a cascade on ..

i'm not trying to discourage you entirely, trying this, but your attempt suffers from:

• pose problem. you'd need to train 2 classifiers, like cow_frontal, and cow_profile (and flip images for the latter)
• texture problem, no cow skin is the same, that's bad.
• not enough positives. images. you'd need like ~100 times more
• -w=24 -h=24, that's probably too small, and the wrong shape(distortion). w=64, h=32 might work much better

also note, there's a pretrained cow detector in opencv_contrib/dpm, also there are pretrained dnn's that can detect cowscows (also, re-training / transfer-learning with cnn's can be done with a surprisingly low amount of images, look it up !)

it's amazing, what kind of things folks are trying to train a cascade on ..

i'm not trying to discourage you entirely, trying this, but your attempt suffers from:

• pose problem. you'd need to train 2 classifiers, like cow_frontal, and cow_profile (and flip images for the latter)
• texture problem, no cow skin is the same, that's bad.bad. terrible. total bummer.
• not enough images. you'd need like ~100 times more
• -w=24 -h=24, that's probably too small, and the wrong shape(distortion). w=64, h=32 might work much better

also note, there's a pretrained cow detector in opencv_contrib/dpm, also there are pretrained dnn's that can detect cows (also, re-training / transfer-learning with cnn's can be done with a surprisingly low amount of images, look it up !)

it's amazing, what kind of things folks are trying to train a cascade on ..

i'm not trying to discourage you entirely, trying this, but your attempt suffers from:

• pose problem. you'd need to train 2 classifiers, like cow_frontal, and cow_profile (and flip images for the latter)
• texture problem, no cow skin is the same, that's bad. terrible. total bummer.
• not enough images. you'd need like ~100 ~ 10 to 100 times more
• -w=24 -h=24, that's probably too small, and the wrong shape(distortion). w=64, h=32 might work much better

also note, there's a pretrained cow detector in opencv_contrib/dpm, also there are pretrained dnn's that can detect cows (also, re-training / transfer-learning with cnn's can be done with a surprisingly low amount of images, look it up !)

it's amazing, what kind of things folks are trying to train a cascade on ..

i'm not trying to discourage you entirely, trying this, but your attempt suffers from:

• pose problem. you'd need to train 2 classifiers, like cow_frontal, and cow_profile (and flip images for the latter)
• texture problem, no cow skin is the same, that's bad. terrible. total bummer.
• not enough images. you'd need like ~ 10 to 100 times more
• -w=24 -h=24, that's probably too small, and the wrong shape(distortion). w=64, h=32 might work much better

also note, there's a pretrained cow detector in opencv_contrib/dpm, also there are pretrained dnn's that can detect cows (also, re-training / transfer-learning with cnn's can be done with a surprisingly low amount of images, look it up !)

it's amazing, what kind of things folks are trying to train a cascade on ..

i'm not trying to discourage you entirely, trying this, but your attempt suffers from:

• pose problem. you'd need to train 2 classifiers, like cow_frontal, and cow_profile (and flip images for the latter)latter, lookup, how this is done for faces)
• texture problem, no cow skin is the same, that's bad. terrible. total bummer.
• not enough images. you'd need like ~ 10 to 100 times more
• -w=24 -h=24, that's probably too small, and the wrong shape(distortion). w=64, h=32 might work much better

also note, there's a pretrained cow detector in opencv_contrib/dpm, also there are pretrained dnn's that can detect cows (also, re-training / transfer-learning with cnn's can be done with a surprisingly low amount of images, look it up !)

My images aren't of that good a quality

it's not bad at all, the problem is more a fundamental / systemic one.

it's amazing, what kind of things folks are trying to train a cascade on ..

i'm not trying to discourage you entirely, trying this, but your attempt suffers from:

• pose problem. you'd need to train 2 classifiers, like cow_frontal, and cow_profile (and flip images for the latter, lookup, how this is done for faces)
• texture problem, no cow skin is the same, that's bad. terrible. total bummer.
• not enough images. you'd need like ~ 10 to 100 times more
• -w=24 -h=24, that's probably too small, and the wrong shape(distortion). w=64, h=32 might work much better

also note, there's a pretrained cow detector in opencv_contrib/dpm, also there are pretrained dnn's that can detect cows (also, (and then, re-training / transfer-learning with cnn's can be done with a surprisingly low amount of images, look it up !)

My images aren't of that good a quality

it's not bad at all, the problem is more a fundamental / systemic one.

it's amazing, what kind of things folks are trying to train a cascade on ..

i'm not trying i don't want to discourage you entirely, trying this, but your attempt suffers from:

• pose problem. you'd need to train 2 classifiers, like cow_frontal, and cow_profile (and flip images for the latter, lookup, how this is done for faces)
• texture problem, no cow skin is the same, that's bad. terrible. total bummer.
• not enough images. you'd need like ~ 10 to 100 times more
• -w=24 -h=24, that's probably too small, and the wrong shape(distortion). w=64, h=32 might work much better

also note, there's a pretrained cow detector in opencv_contrib/dpm, also there are pretrained dnn's that can detect cows (and then, re-training / transfer-learning with cnn's can be done with a surprisingly low amount of images, look it up !)

My images aren't of that good a quality

it's not bad at all, the problem is more a fundamental / systemic one.

it's amazing, what kind of things folks are trying to train a cascade on ..

i don't want to discourage you entirely, trying this, but your attempt suffers from:

• pose problem. you'd need to train 2 classifiers, like cow_frontal, and cow_profile (and flip images for the latter, lookup, how this is done for faces)
• texture problem, no cow skin is the same, that's bad. terrible. total bummer.
• not enough images. you'd need like ~ 10 to 100 times more
• -w=24 -h=24, that's probably too small, and the wrong shape(distortion). w=64, h=32 might work much better

also note, there's a pretrained cow detector in opencv_contrib/dpm, also there are pretrained dnn's that can detect cows (and then, re-training / transfer-learning with (ssd or yolo) cnn's can be done with a surprisingly low amount of images, look it up !)

My images aren't of that good a quality

it's not bad at all, the problem is more a fundamental / systemic one.