This comment has been removed by the author.
This one is better than "More stringent pedestrian detection by Deep learning verification" and "More stringent pedestrian detection by HOG verification". Do you use OpenCV to achieve such result?
Detecting pedestrian using Deep learning is recorded high rate false positive, but not sensitive variable pedestrian pose. and processing speed is not good.Hog is fast, but to detect typical appearance when training data pose.So, I can not say yet, which way is better..Thank you.
can you share this source code! Thank you