Eugene Zatepyakin
Eugene Zatepyakin
this is not that easy as just using NPD instead of pixel diffs. you should take into account that response values will be different so you should select threshold more...
in my implementation i'm generating random splits with 2 thresholds one negative and one positive. threshold values are in range -127 to 127. so for each split i try 10...
just wondering how many negative images you are using? and what is the average image size (width x height)?
i think 5 landmarks still can do a good job and using less landmarks will produce more compact model file in the end. in my tests i see that LBF...
well as far as u see there is no method to pase it internally but if u know the structure of the file its quite easy just to push correct...
for performance reasons jsfeat uses simple average of 4 pixels during pyr_down. if u are interested in very precise resizing during down sampling u can use jsfeat resize function.
there is no such method available at the moment. but u can always write/add/port it in ;)
hi thanx for your interest. jsfeat impl of HAAR cascades is based on an old version of trainer from OpenCV and it is compatible to stump based classifiers. Classifiers for...
we should ask the author @gaborpapp
this is handled on usb driver level. as far as i remember there was no way to handle it inside PS3EYEDriver