nouamanetazi HF staff commited on
Commit
6370852
1 Parent(s): d142740

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +24 -0
README.md ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - video-prediction
4
+ - moving-mnist
5
+ - video-to-video
6
+ license: cc0-1.0
7
+ ---
8
+ ## Tensorflow Keras Implementation of Next-Frame Video Prediction with Convolutional LSTMs 📽️
9
+
10
+ This repo contains the models and the notebook [on How to build and train a convolutional LSTM model for next-frame video prediction](https://keras.io/examples/vision/conv_lstm/).
11
+
12
+ Full credits to [Amogh Joshi](https://github.com/amogh7joshi)
13
+
14
+ ## Background Information
15
+ The [Convolutional LSTM](https://papers.nips.cc/paper/2015/file/07563a3fe3bbe7e3ba84431ad9d055af-Paper.pdf) architectures bring together time series processing and computer vision by introducing a convolutional recurrent cell in a LSTM layer. This model uses the Convolutional LSTMs in an application to next-frame prediction, the process of predicting what video frames come next given a series of past frames.
16
+
17
+ ![preview](https://keras.io/img/examples/vision/conv_lstm/conv_lstm_13_0.png)
18
+
19
+ ## Training Dataset
20
+ This model was trained on the [Moving MNIST dataset](http://www.cs.toronto.edu/~nitish/unsupervised_video/).
21
+
22
+ For next-frame prediction, our model will be using a previous frame, which we'll call `f_n`, to predict a new frame, called `f_(n + 1)`. To allow the model to create these predictions, we'll need to process the data such that we have "shifted" inputs and outputs, where the input data is frame `x_n`, being used to predict frame `y_(n + 1)`.
23
+
24
+ ![result](https://i.imgur.com/UYMTsw7.gif)