Bookmark and Share

Computer Vision Metrics: Chapter Eight

Register or sign in to access the Embedded Vision Academy's free technical training content.

The training materials provided by the Embedded Vision Academy are offered free of charge to everyone. All we ask in return is that you register, and tell us a little about yourself so that we can understand a bit about our audience. As detailed in our Privacy Policy, we will not share your registration information, nor contact you, except with your consent.

Registration is free and takes less than one minute. Click here to register, and get full access to the Embedded Vision Academy's unique technical training content.

If you've already registered, click here to sign in.

See a sample of this page's content below:


Bibliography references are set off with brackets, i.e. "[XXX]". For the corresponding bibliography entries, please click here.


Vision Pipelines and Optimizations

“More speed, less haste . . . ”
—Treebeard, Lord of the Rings

This chapter explores some hypothetical computer vision pipeline designs to understand HW/SW design alternatives and optimizations. Instead of looking at isolated computer vision algorithms, this chapter ties together many concepts into complete vision pipelines. Vision pipelines are sketched out for a few example applications to illustrate the use of different methods. Example applications include object recognition using shape and color for automobiles, face detection and emotion detection using local features, image classification using global features, and augmented reality. The examples have been chosen to illustrate the use of different families of feature description metrics within the Vision Metrics Taxonomy presented in Chapter 5. Alternative optimizations at each stage of the vision pipeline are explored. For example, we consider which vision algorithms run better on a CPU versus a GPU, and discuss how data transfer time between compute units and memory affects performance.

Note: The hypothetical examples in this chapter are sometimes sketchy, not intended to be complete. Rather, the intention is to explore design alternatives. Design choices are made in the examples for illustration only; other, equally valid design choices could be made to build working systems. The reader is encouraged to analyze the examples to find weaknesses and alternatives. If the reader can improve the examples, we have succeeded.

This chapter addresses the following major topics, in this order:

  1. General design concepts for optimization across the SOC (CPU, GPU, memory).
  2. Four hypothetical vision pipeline designs using different descriptor methods.
  3. Overview of SW optimization resources and specific optimization techniques.

Stages, Operations, and Resources

A computer vision solution can be implemented into a pipeline of stages, as shown in Figure 8-1. In a pipeline, both parallel and sequential operations take place simultaneously. By using all available compute resources in the optimal manner, performance can be maximized for speed, power, and memory efficiency.


Figure 8-1. Hypothetical assignment of vision pipeline stages to operations and to compute resources. Depending on the actual...