Upload README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pretty_name: HUD UI Validation Pilot
|
| 3 |
+
license: other
|
| 4 |
+
task_categories:
|
| 5 |
+
- image-classification
|
| 6 |
+
tags:
|
| 7 |
+
- gameplay
|
| 8 |
+
- hud
|
| 9 |
+
- ui
|
| 10 |
+
- validation
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# HUD/UI annotation validation pilot
|
| 14 |
+
|
| 15 |
+
This public artifact compares 10 gameplay screenshots across four columns:
|
| 16 |
+
|
| 17 |
+
1. raw frame;
|
| 18 |
+
2. GPT-5.6 Sol X-High final reference;
|
| 19 |
+
3. GPT-5.6 Terra High refined annotation;
|
| 20 |
+
4. confidence-routed final annotation.
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
|
| 24 |
+
Files:
|
| 25 |
+
|
| 26 |
+
- [`analysis.md`](analysis.md): aggregate metrics and per-game error table;
|
| 27 |
+
- [`per_sample_metrics.csv`](per_sample_metrics.csv): machine-readable sample metrics;
|
| 28 |
+
- [`overlay_comparison_contact_sheet.jpg`](overlay_comparison_contact_sheet.jpg): full comparison sheet.
|
| 29 |
+
|
| 30 |
+
The 0--100 confidence value is a conservative pipeline routing score, not a
|
| 31 |
+
calibrated probability of correctness. On this 10-sample pilot, pure Terra High
|
| 32 |
+
element F1 is 0.7755 and mean slot Jaccard is 0.6875. The sample is too small for
|
| 33 |
+
a final model-selection conclusion.
|