cvtechniques
/

JC-Traffic-Sign-Detection

Model card Files Files and versions

xet

Community

jahmesuw commited on Mar 18

Commit

2cb01e6

verified ·

1 Parent(s): ea71aeb

Update README.md

Browse files

Files changed (1) hide show

README.md +6 -6

README.md CHANGED Viewed

@@ -109,7 +109,7 @@ All three models perform similarly across evaluation metrics, with only small di
 | YOLOv11s | 0.894 | 0.973 | 0.948 | 0.777 |
 | YOLOv26s | 0.919 | 0.922 | 0.951 | 0.781 |
-![Model Comparison](https://huggingface.co/datasets/cvtechniques/JC-Traffic-Sign-Detection/resolve/main/results/model_comparison.png)
 The three YOLO models show very similar overall performance, with only small differences across evaluation metrics.
@@ -118,7 +118,7 @@ YOLOv11 performs slightly better overall, while YOLOv26 achieves nearly identica
 YOLOv8 performs marginally lower but remains within a small margin of error.
 ### F1 Curve
-![F1_Curve](https://huggingface.co/datasets/cvtechniques/JC-Traffic-Sign-Detection/resolve/main/results/f1_curve.png)
 The F1-confidence curves show strong balance between precision and recall across all three models.
 The peak F1 score for all classes reaches approximately 0.92, indicating high overall detection performance, with all three models performing very similarly.
@@ -126,7 +126,7 @@ Most classes maintain high F1 scores across confidence thresholds. However, the
 This likely reflects fewer training examples for these classes and the visual similarity between speed limit signs, where only the number differs, making them more difficult for the model to distinguish.
 ### Confusion Matrix
-![matrix](https://huggingface.co/datasets/cvtechniques/JC-Traffic-Sign-Detection/resolve/main/results/matrix.png)
 The confusion matrix show strong classification performance across all three models, with most predictions showing up along the diagonal.
 This indicates that the majority of traffic sign classes are correctly classified with very few misclassifications.
@@ -134,7 +134,7 @@ The three models display very similar patterns, suggesting consistent class reco
 Minor misclassifications occur between some speed limit classes, where predictions occasionally fall into adjacent speed categories.
 ### Sample Images
-![images](https://huggingface.co/datasets/cvtechniques/JC-Traffic-Sign-Detection/resolve/main/results/sample_images.png)
 The sample predictions demonstrate that all three models can accurately detect and localize traffic signs across a variety of road environments.
 Bounding boxes are placed correctly around most signs, and detection performance is generally consistent across the three models, although YOLOv26 occasionally struggles to detect the speedLimit55 sign compared to the other architectures.
@@ -145,7 +145,7 @@ Overall, the predictions show strong and reliable detection performance across d
 ### Upscaling Comparison (1080p, 1440p, 2160p)
 <video width="800" controls>
-  <source src="https://huggingface.co/datasets/cvtechniques/JC-Traffic-Sign-Detection/resolve/main/results/cv_upscaling.mov" type="video/mp4">
 </video>
 The upscaling video uses a 0.25 confidence threshold on YOLOv26 and shows that increasing the image resolution can help the model detect smaller or distant traffic signs by making important visual features more visible.
@@ -155,7 +155,7 @@ However, some false positives or misclassifications may still occur when the ori
 ### v11 (TTA) vs v26 (non-TTA)
 <video width="800" controls>
-  <source src="https://huggingface.co/datasets/cvtechniques/JC-Traffic-Sign-Detection/resolve/main/results/cv_v11%26v26.mov" type="video/mp4">
 </video>
 This video compares detection performance between YOLOv11 using Test Time Augmentation (TTA) and YOLOv26 without TTA.

 | YOLOv11s | 0.894 | 0.973 | 0.948 | 0.777 |
 | YOLOv26s | 0.919 | 0.922 | 0.951 | 0.781 |
+![Model Comparison](https://huggingface.co/cvtechniques/JC-Traffic-Sign-Detection/resolve/main/results/model_comparison.png)
 The three YOLO models show very similar overall performance, with only small differences across evaluation metrics.
 YOLOv8 performs marginally lower but remains within a small margin of error.
 ### F1 Curve
+![F1_Curve](https://huggingface.co/cvtechniques/JC-Traffic-Sign-Detection/resolve/main/results/f1_curve.png)
 The F1-confidence curves show strong balance between precision and recall across all three models.
 The peak F1 score for all classes reaches approximately 0.92, indicating high overall detection performance, with all three models performing very similarly.
 This likely reflects fewer training examples for these classes and the visual similarity between speed limit signs, where only the number differs, making them more difficult for the model to distinguish.
 ### Confusion Matrix
+![matrix](https://huggingface.co/cvtechniques/JC-Traffic-Sign-Detection/resolve/main/results/matrix.png)
 The confusion matrix show strong classification performance across all three models, with most predictions showing up along the diagonal.
 This indicates that the majority of traffic sign classes are correctly classified with very few misclassifications.
 Minor misclassifications occur between some speed limit classes, where predictions occasionally fall into adjacent speed categories.
 ### Sample Images
+![images](https://huggingface.co/cvtechniques/JC-Traffic-Sign-Detection/resolve/main/results/sample_images.png)
 The sample predictions demonstrate that all three models can accurately detect and localize traffic signs across a variety of road environments.
 Bounding boxes are placed correctly around most signs, and detection performance is generally consistent across the three models, although YOLOv26 occasionally struggles to detect the speedLimit55 sign compared to the other architectures.
 ### Upscaling Comparison (1080p, 1440p, 2160p)
 <video width="800" controls>
+  <source src="https://huggingface.co/cvtechniques/JC-Traffic-Sign-Detection/resolve/main/results/cv_upscaling.mov" type="video/mp4">
 </video>
 The upscaling video uses a 0.25 confidence threshold on YOLOv26 and shows that increasing the image resolution can help the model detect smaller or distant traffic signs by making important visual features more visible.
 ### v11 (TTA) vs v26 (non-TTA)
 <video width="800" controls>
+  <source src="https://huggingface.co/cvtechniques/JC-Traffic-Sign-Detection/resolve/main/results/cv_v11%26v26.mov" type="video/mp4">
 </video>
 This video compares detection performance between YOLOv11 using Test Time Augmentation (TTA) and YOLOv26 without TTA.