Delauney's profile - activity

So this isn't unexpectedly slow?

I'm training a cascade classifier now and it's on the 6th stage and has been there for more than 48 hours so far. It's slowly working through the negatives. I'm wondering if the resolution of the negatives is causing this? They are large-ish images (roughly desktop background resolution). I noticed that using cascade classifiers is much faster with smaller images so I can only imagine that training with large images slows it down as well. Any experience with this?

So what is the ideal resolution for negatives?

I'm not sure what the original cause for this error was but ultimately I recreated the negatives.txt file using full file paths and also did not use -featureType switch.

I think you have to have mp4 codec installed in your OS. Make sure your environment can open/understand MP4 before OpenCV can understand it. http://opensuse-guide.org/codecs.php

Hello everyone. An internet search suggested that this was a bug that was patched in the past so I have no idea what I'm doing wrong to recreate this bug. I've tried it in 2.4.11 and 3.0.0. Running on Linux Mint 17.2 64-bit.

Here's the full command and output:

user@linux ~/file/path $ opencv_traincascade -data ~/file/path/ -vec cards.vec -bg negatives.txt -numPos 500 -numNeg -500 -numStages 2 -w 53 -h 74 -featureType LBP
OpenCV Error: Assertion failed (_maxSampleCount > 0) in init, file /home/user/Documents/opencv-3.0.0/apps/traincascade/lbpfeatures.cpp, line 17
terminate called after throwing an instance of 'cv::Exception'
  what():  /home/user/Documents/opencv-3.0.0/apps/traincascade/lbpfeatures.cpp:17: error: (-215) _maxSampleCount > 0 in function init

Aborted

I have 790 positive samples to use and 40,000 negative samples so I know I'm not running out. What could cause this? Is it possibly a bug in my negatives.txt? I've followed several different guides/tutorials from the internet, all to no avail.