Revision history [back]

imho, you're going all the wrong way here.

1. why the black bg ? it's probably not what your detection scenario looks like. the training has to learn to discriminate object and background features, you're defeating it
2. this needs a ton of REAL images, you can't succsessfully synthesize 5000 images from 5 (so, drop the perl script, and acquire as many images with real bg as you can.)

imho, you're going all the wrong way here.2 problems here:

1. why the black bg ? it's probably not what your detection scenario looks like. the training has to learn to discriminate object and background features, you're defeating it
2. this needs a ton of REAL images, you can't succsessfully synthesize 5000 images from 5 (so, drop the perl script, and acquire as many images with real bg as you can.)

imho, 2 3 problems here:

1. why the black bg ? it's probably not what your detection scenario looks like. the training has to learn to discriminate object and background features, you're defeating it
2. this needs a ton of REAL images, you can't succsessfully synthesize 5000 images from 5 (so, drop the perl script, and acquire as many images with real bg as you can.)
3. please use official tutorials, not outdated blog posts

imho, 3 problems here:

1. here: 1. please use official tutorials, not outdated blog posts 2. why the black bg ? it's probably not what your detection scenario looks like. the training has to learn to discriminate object and background features, you're defeating it
2. it 3. this needs a ton of REAL images, you can't succsessfully synthesize 5000 images from 5 (so, drop the perl script, and acquire as many images with real bg as you can.)
3. please use official tutorials, not outdated blog posts

imho, 3 problems here: 1. here:

1. please use official tutorials, not outdated blog posts 2. posts
2. why the black bg ? it's probably not what your detection scenario looks like. the training has to learn to discriminate object and background features, you're defeating it 3. it
3. this needs a ton of REAL images, you can't succsessfully synthesize 5000 images from 5 (so, drop the perl script, and acquire as many images with real bg as you can.)

imho, 3 problems here:

1. please use official tutorials, not outdated blog posts
2. why the black bg ? it's probably not what your detection scenario looks like. the training has to learn to discriminate object and background features, you're defeating it
3. this needs a ton of REAL images, you can't train with 18 positives only, and also you can't succsessfully synthesize 5000 images from 5 (so, drop the perl script, and acquire as many images with real bg as you can.)

imho, 3 problems here:

1. please use official tutorials, not outdated blog posts
2. why the black bg ? it's probably not what your detection scenario looks like. the training has to learn to discriminate object and background features, you're defeating it
3. this needs a ton of REAL images, you can't train with 18 positives only, and also you can't succsessfully successfully synthesize 5000 images from 5 (so, drop the perl script, and acquire as many images with real bg as you can.)