KaiyangZhou / pytorch-vsumm-reinforce

Unsupervised video summarization with deep reinforcement learning (AAAI'18)
MIT License
471 stars 150 forks source link

Can someone explain the reward function? #63

Open prathamm03 opened 3 years ago

prathamm03 commented 3 years ago

As far as I understand, diversity reward is based on the dissimilarity between the selected frames, that dissimilarity might be based on features of the image, but I can't understand how this representative reward works, can anyone make it clear?