Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix bug that creates an empty output image #148

Merged
merged 1 commit into from
Sep 11, 2023

Conversation

pataxis
Copy link
Contributor

@pataxis pataxis commented Sep 11, 2023

Describe your changes

138 reported that the cropped images saved are black. The issue is that the pre-processing was not applied to the high-resolution stream. On the way, more bugs were fixed.

object-detector:

  • Pre-processing of the high resolution frame is added again to fix 138
  • Fixed issue when calculating crop size of the HD image that was causing crops to be slightly off

object-detector-cv25:

  • All issues found in the other object-detection example are applied also on the CV25 version
  • Fixed padding function, now it copies all the channels.
  • Creating an extra tensor to hold the low resolution image before padding.
  • Increased the resolution of the HD image to 1280x720 from 640x360. It was previously lowered from 1920x1080 because not all the CV25 products were supporting 1080p, but it should be the maximum resolution supported by all CV25 devices.

Also, some refactoring of the variable was done to give some sense to their names.

Issue ticket number and link

@pataxis pataxis requested a review from a team as a code owner September 11, 2023 16:11
Bug in object-detection* examples.
Fixes #138
@pataxis pataxis force-pushed the fix/object-detection-black-image branch from 448c274 to a844dd1 Compare September 11, 2023 16:19
@pataxis pataxis merged commit 2349aa5 into main Sep 11, 2023
7 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Development

Successfully merging this pull request may close these issues.

Black cropped image is output when detect object in object detection
2 participants