abhisheks008 · Adhavan1801 · May 29, 2026 · May 29, 2026 · May 29, 2026 · May 29, 2026
diff --git a/Multi-Modal Road Damage Detection/.gitignore b/Multi-Modal Road Damage Detection/.gitignore
diff --git a/Multi-Modal Road Damage Detection/Dataset/README.md b/Multi-Modal Road Damage Detection/Dataset/README.md
@@ -0,0 +1,44 @@
+# Dataset — RDD2022 (Road Damage Dataset 2022)
+
+## Source
+**Kaggle:** https://www.kaggle.com/datasets/aliabdelmenam/rdd-2022
+
+## Description
+RDD2022 is a large-scale multi-country road damage dataset used for
+automatic road damage detection and classification.
+
+- **Size:** ~10.3 GB
+- **Total Images:** 38,385
+- **Countries covered:** Japan, India, Czech Republic, Norway, United States, China (Drone & MotorBike)
+- **Damage Categories:**
+  | Label | Description |
+  |-------|-------------|
+  | D00   | Longitudinal Cracks |
+  | D10   | Transverse Cracks |
+  | D20   | Alligator Cracks |
+  | D40   | Potholes |
+
+## Dataset Structure
+```
+   RDD_SPLIT/
+    ├── train/
+    ├── val/
+    └── test/
+        ├── images/    ← .jpg files (country-prefixed filenames)
+        │   ├── China_Drone_*.jpg
+        │   ├── China_MotorBike_*.jpg
+        │   ├── Czech_*.jpg
+        │   ├── India_*.jpg
+        │   ├── Japan_*.jpg
+        │   ├── Norway_*.jpg
+        │   └── United_States_*.jpg
+        │
+        └── labels/    ← YOLO format .txt annotation files
+            ├── China_Drone_*.txt
+            ├── China_MotorBike_*.txt
+            ├── Czech_*.txt
+            ├── India_*.txt
+            ├── Japan_*.txt
+            ├── Norway_*.txt
+            └── United_States_*.txt
+```
diff --git a/Multi-Modal Road Damage Detection/Images/.gitkeep b/Multi-Modal Road Damage Detection/Images/.gitkeep
@@ -0,0 +1 @@
+
diff --git a/Multi-Modal Road Damage Detection/Images/class_distribution.png b/Multi-Modal Road Damage Detection/Images/class_distribution.png
diff --git a/Multi-Modal Road Damage Detection/Images/country_wise_damage_distribution.png b/Multi-Modal Road Damage Detection/Images/country_wise_damage_distribution.png
diff --git a/Multi-Modal Road Damage Detection/Images/efficientnet_training_curves.png b/Multi-Modal Road Damage Detection/Images/efficientnet_training_curves.png
diff --git a/Multi-Modal Road Damage Detection/Images/gradcam_comparison_grid.png b/Multi-Modal Road Damage Detection/Images/gradcam_comparison_grid.png
diff --git a/Multi-Modal Road Damage Detection/Images/gradcam_efficientnet_b0.png b/Multi-Modal Road Damage Detection/Images/gradcam_efficientnet_b0.png
diff --git a/Multi-Modal Road Damage Detection/Images/gradcam_resnet50.png b/Multi-Modal Road Damage Detection/Images/gradcam_resnet50.png
diff --git a/Multi-Modal Road Damage Detection/Images/gradcam_vit_b16.png b/Multi-Modal Road Damage Detection/Images/gradcam_vit_b16.png
diff --git a/Multi-Modal Road Damage Detection/Images/resnet50_training_curves.png b/Multi-Modal Road Damage Detection/Images/resnet50_training_curves.png
diff --git a/Multi-Modal Road Damage Detection/Images/train_samples.png b/Multi-Modal Road Damage Detection/Images/train_samples.png
diff --git a/Multi-Modal Road Damage Detection/Images/vit_training_curve.png b/Multi-Modal Road Damage Detection/Images/vit_training_curve.png
diff --git a/Multi-Modal Road Damage Detection/Model/.gitkeep b/Multi-Modal Road Damage Detection/Model/.gitkeep
@@ -0,0 +1 @@
+
diff --git a/Multi-Modal Road Damage Detection/Model/Road_Damage_Detection.ipynb b/Multi-Modal Road Damage Detection/Model/Road_Damage_Detection.ipynb
diff --git a/Multi-Modal Road Damage Detection/Model/model_results_summary.csv b/Multi-Modal Road Damage Detection/Model/model_results_summary.csv
@@ -0,0 +1,5 @@
+model,accuracy,precision,recall,f1,latency_ms
+EfficientNet_B0,0.7131,0.7137,0.7131,0.7131,8.4771
+ResNet50,0.7586,0.7572,0.7586,0.7573,8.6351
+YOLOv8n,0.5659,0.6196,0.529,0.5707,7.5172
+ViT_B16,0.7342,0.7362,0.7342,0.7275,12.9514