Revision history [back]

Hi, last few days, I'm trying to train my hog descriptor for pedestrian recognition. I ran into with the following problem: training is OK but when i test the detector on negative images, it detect 1 person in the center of the image... Does anyone have any idea why this is happening?

Hi,

for the last few days, I'm trying to train my hog descriptor for pedestrian recognition. I ran into with the following problem: training is OK but when i I test the detector on negative images, it detect detects 1 person in the center of the image...

Does anyone have any idea why this is happening?