[FR] Draw bounding box/segmentation mask heatmap of a dataset

[suvojit-0x55aa]

Proposal Summary

Option to draw bounding box/segmentation mask heatmap of a dataset, and see how the the objects are distributed spatially.

Motivation

• What is the use case for this feature?

• This feature can be used to understood the places where the objects generally lie and visualize the trends

• Why is this use case valuable to support for FiftyOne users in general?

• Fiftyone already has the necessary query capabilities built in to support this.

• Why is this use case valuable to support for your project(s) or organization?

It's easier to identify the regions in which objects are

• Why is it currently difficult to achieve this use case? (please be as specific as possible about why related FiftyOne features and components are insufficient)

Has to be done manually.

What areas of FiftyOne does this feature affect?

• [ ] App: FiftyOne application
• [x] Core: Core fiftyone Python library
• [ ] Server: FiftyOne server

Details

(Use this section to include any additional information about the feature. If you have a proposal for how to implement this feature, please include it here.)

Willingness to contribute

The FiftyOne Community encourages new feature contributions. Would you or another member of your organization be willing to contribute an implementation of this feature?

• [ ] Yes. I can contribute this feature independently.
• [ ] Yes. I would be willing to contribute this feature with guidance from the FiftyOne community.
• [x] No. I cannot contribute this feature at this time.

[brimoor]

Hi @suvojit-0x55aa 👋

This sounds like a useful feature to have. My proposal would be that this starts as a method in a new module within the fiftyone.utils subpackage like this:

def visualize_object_locations(sample_collection, label_field, ...):
    ...

which would return the heatmap data and/or plot it directly. If features like this become heavily used, we could support generating + viewing such heatmaps directly in the App.

Would you be able to take a shot at contributing a PR that gets this feature off the ground? 😄

[daniel-merrick]

Was this ever created?