liveBook · Manning

Video

As explained in Deep Learning with Python, Third Edition

Data in which each example is a sequence of image frames, typically stored as a rank-5 tensor for a batch.

Video data represents each example as a sequence of image frames. A batch is stored as a rank-5 tensor with shape (samples, frames, height, width, channels). section 2.2.8

Representing a batch of videos

For a batch of video clips, read the axes as follows: samples selects the clip, frames selects an image within it, height and width locate pixels, and channels store each pixel’s values. section 2.2.8

Use this rule of thumb:

  • Images use rank-4 tensors: (samples, height, width, channels); video adds a frames axis. section 2.2.8

How Deep Learning with Python, Third Edition covers it

How deep it goes
Defines the rank-5 shape and axis meanings, but stops short of explaining video-processing operations or model architectures.
Examples
Uses a generic batch of video clips organized by samples, frames, height, width, and channels.
What you can do after
Evaluate whether video data has been arranged with the expected five tensor axes.
How it connects
Connects video with image data, sequence data, and tensors through nested frames, pixels, and channels.
Cover of Deep Learning with Python, Third Edition
liveBookmanning.com
Read Deep Learning with Python, Third Edition on liveBook Get the book at manning.com

Also explained in

Compare how 5 books explain video →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Deep Learning with Python, Third Edition. The section links point to the parts of the book it draws on.