[ITEP-81197] Publish to scene topics when there are no detections

[jakubsikorski]

📝 Description

ITEP-81197

Detections are no longer hanging on the scene UI and disappear as no detections occur.

✨ Type of Change

Select the type of change your PR introduces:

• 🐞 Bug fix – Non-breaking change which fixes an issue
• 🚀 New feature – Non-breaking change which adds functionality
• 🔨 Refactor – Non-breaking change which refactors the code base
• 💥 Breaking change – Changes that break existing functionality
• 📚 Documentation update
• 🔒 Security update
• 🧪 Tests
• 🚂 CI

🧪 Testing Scenarios

Describe how the changes were tested and how reviewers can test them too:

• ✅ Tested manually
• 🤖 Ran automated end-to-end tests

✅ Checklist

Before submitting the PR, ensure the following:

• 🔍 PR title is clear and descriptive
• 📝 For internal contributors: If applicable, include the JIRA ticket number (e.g., ITEP-123456) in the PR title. Do not include full URLs
• 💬 I have commented my code, especially in hard-to-understand areas
• 📄 I have made corresponding changes to the documentation
• ✅ I have added tests that prove my fix is effective or my feature works

[Copilot]

Pull Request Overview

This PR fixes an issue where detection objects were persisting in the scene UI when no new detections occurred. The change ensures that when a message contains no detection types, empty arrays are explicitly added for all tracked types before processing, which allows the tracker to properly clear stale detections.

• Adds logic to handle messages with no detections by populating empty arrays for all tracked detection types

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The variable detection_types is reassigned from an empty list to a populated list, but this reassignment only affects the local scope. If detection_types is used elsewhere in the function after this block, the reassignment could lead to unexpected behavior. Consider using a different variable name (e.g., tracked_types) for the list of tracker keys to avoid confusion and maintain clarity about which detection types are being referenced.

Suggested change

	detection_types = list(scene.tracker.trackers.keys())
	for dtype in detection_types:
	tracked_types = list(scene.tracker.trackers.keys())
	for dtype in tracked_types:

[jakubsikorski]

Changing it this way would be confusing. detection_types can be (line 400) a reference to the dictionary keys, not a copy. Modifying jdata (referenced keys) will also modify detection_types because it's a dictionary view.

The explicit reassignment makes this side effect visible without requiring readers to understand dictionary view semantics.

[rawatts10]

Verified working, so approved. I don't really like the idea of caching past tracked_types, but because we publish per-class topics it may be necessary. We should revisit this in the future, particularly since there are other edge cases that should be considered (e.g. cameras going down -- if no frames are coming from the camera then the last detection stays in the scene). This is something we can take up with TTL and persistence on the scene.