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

data format #47

Closed
namemute opened this issue Nov 27, 2024 · 2 comments
Closed

data format #47

namemute opened this issue Nov 27, 2024 · 2 comments

Comments

@namemute
Copy link

Hello,I want to know the specific meaning of the data in npy under the mask folder. For example, What does each line and column represent in occluded_reid/masks/pifpaf_maskrcnn_filtering/whole_body_images/002_02.npy?
)D{$19OMHURQ81PC%@ILVZP

@VlSomers
Copy link
Owner

Hi @namemute, you can have a look at the function to read these masks here to have more information:

def read_masks(masks_path):

Basically, this is an array of size H x W x N, that contains N masks of size HxW with values in between 0 and 1. These masks were generated using the pif and paf fields returned by OpenPifPaf, with N=36. You can have a look at this script to see how they are generated: https://github.com/VlSomers/bpbreid/blob/main/torchreid/scripts/get_labels.py. You can also simply visualize these masks as heatmaps on top of their corresponding image to get a better sense of their meaning. Please also have a look at the conversation of this PR: #18

@namemute
Copy link
Author

namemute commented Dec 3, 2024

Hi @namemute, you can have a look at the function to read these masks here to have more information:

def read_masks(masks_path):

Basically, this is an array of size H x W x N, that contains N masks of size HxW with values in between 0 and 1. These masks were generated using the pif and paf fields returned by OpenPifPaf, with N=36. You can have a look at this script to see how they are generated: https://github.com/VlSomers/bpbreid/blob/main/torchreid/scripts/get_labels.py. You can also simply visualize these masks as heatmaps on top of their corresponding image to get a better sense of their meaning. Please also have a look at the conversation of this PR: #18

ok,thanks

@namemute namemute closed this as completed Dec 3, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants