Need:
- With the ever-growth large-scale video on the mobile phone, so what would everyone get from these video? There is many videos contain something very interesting like a short comedy video. So if someone find something interesting in the video and want know more about it, they could not search it in the Internet and find the information after watching this video due the poor memory. So if the Advertiser has put some advertisements in the video ahead of time, it'll be is more convenient for the user to G Et some information. That's very useful for the advertisers and the users.
- There is many videos in the users ' phone. Maybe most of them is meaningful time mark. So someone want to look for some useful tools to tagging the meaningful object or want to know the object information. Then our video tagging systems would be very efficient
Approach:
- The video tagging project can divided into and steps. The first one is the key frame localization. The second one is the object classification or object detection.
- The key frame localization can be realized by some conventional method like the HOG features split or some other method. This was a litter challenge because there is no very efficient the "to get" really accuracy key frame. And I think it is a program optimization problem.
- The object classification can be realized by the deep convolutional neural network classifier or some other deep learning State-of-the-arts method. The problem is the labels may was not enough. So it can be a problem.
Benefit:
- Everyone can be convenient to get some merchandise information by the tagged video which are processed by the mobile end AP Plication.
- Some people would summarize the meaningful moments and find Some meaningful object.
Competitors:
There a video tagging system which have been released in the Internet after my survey. The Website name is "Clarifai". They can tag the video and get the object temporal information. And the classification accuracy is very high. So it's our main competitor.
10/18/2015
Fuchen Long
Video tagging systems based on Dnns
