While learning action recognition we need to exploit both global and local temporal features. But cant understand the difference between the two??

More Ashish Karel's questions See All
Similar questions and discussions