Skip to content

Latest commit

 

History

History
16 lines (9 loc) · 474 Bytes

README.md

File metadata and controls

16 lines (9 loc) · 474 Bytes

Video-Emmbeddings

Unsupervised Learning on videos using TensorFlow 2.0, Keras OpenCV is used only for initial preprocessing.

Converting a video (.mp4 format) in to say 256 column vector. So that it can be used for ad reccomendations.

Please see the file Main.ipynb file.

Data used : Fun2.mp4

Coming Soon : Using LSTMs only

Refereces : 1. https://arxiv.org/pdf/1502.04681.pdf

        2. https://www.kaggle.com/valkling/how-to-teach-an-ai-to-dance