Skip to content

Latest commit

 

History

History
4 lines (4 loc) · 640 Bytes

README.md

File metadata and controls

4 lines (4 loc) · 640 Bytes

video-classification

Our idea focuses on continuous video classification using a convolutional neural network(CNN).

We would be classifying advertisements from live streaming(more specifically sports). Videos have both spatial and temporal features but we have only taken spatial features into consideration.

What we see affects our minds a lot more than we actually realize and most advertisements not only waste our time but also affect our brain in a detrimental way. Our solution is to mute your device whenever our model realizes that an ad is playing. For convenience we are considering your laptop as the streaming device.