Skip to content
This repository has been archived by the owner on Sep 18, 2018. It is now read-only.

Lab 3.3 Feature Points (Matching) #5

Closed
6 tasks
rgolovanov opened this issue Nov 12, 2017 · 0 comments
Closed
6 tasks

Lab 3.3 Feature Points (Matching) #5

rgolovanov opened this issue Nov 12, 2017 · 0 comments
Labels

Comments

@rgolovanov
Copy link
Contributor

Preparing

Follow recommendations from previous task.

Task

This is 3nd part of the Feature Points topic:

  • Detecting
  • Describing
  • Matching

It is required to implement application which creates interactive panoramas image.

  1. Work in your private repository created in previous task.

  2. Application shall support panorama building mode:

    1. Handle "space" key to enter panoramas mode.
    2. In panorama mode only two windows shall be shown:
      • Current window showing image from web camera.
      • Panoramic window showing merged images.
    3. Once "space" key pressed again the current image shall be captured and used for panorama building.
    4. When the first image was captured then just find feature points (don't visualize them).
    5. When next image was captured then find all feature points, find intersections, add rest points into database of features related to all captured images.
    6. Handle "Ctrl + Space" to exit from panorama mode.
  3. Find best parameters for detectors and matchers to solve the problem.

Check-list

  • Code is compiling in debug and release configuration without any errors and warnings in your code.
  • All files passes cpplint check.
  • All methods, classes, structures have comments.
  • Debugging text is satisfy requirements.
  • Windows and theirs elements (trackbars) are satisfy requirements.
  • Windows handles user interaction correctly (mouse and keyboard actions).
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
Projects
None yet
Development

No branches or pull requests

1 participant