Additional notes on RAF-DB

kode-git · May 13, 2022 · 36efa7a · 36efa7a
1 parent 88c931a
commit 36efa7a
Showing 1 changed file with 26 additions and 17 deletions.
diff --git a/ViT_Face_Emotion_Recognition.ipynb b/ViT_Face_Emotion_Recognition.ipynb
@@ -66,7 +66,7 @@
     },
     {
       "cell_type": "code",
-      "execution_count": 5,
+      "execution_count": null,
       "metadata": {
         "colab": {
           "base_uri": "https://localhost:8080/"
@@ -117,7 +117,7 @@
         "id": "8wNJEIuURXEb",
         "outputId": "a90ac5ac-513d-4257-bddb-8f641ac5313c"
       },
-      "execution_count": 6,
+      "execution_count": null,
       "outputs": [
         {
           "output_type": "stream",
@@ -130,7 +130,7 @@
     },
     {
       "cell_type": "code",
-      "execution_count": 7,
+      "execution_count": null,
       "metadata": {
         "id": "66hzxAClAFAY"
       },
@@ -201,7 +201,7 @@
     },
     {
       "cell_type": "code",
-      "execution_count": 8,
+      "execution_count": null,
       "metadata": {
         "colab": {
           "base_uri": "https://localhost:8080/"
@@ -251,7 +251,7 @@
     },
     {
       "cell_type": "code",
-      "execution_count": 9,
+      "execution_count": null,
       "metadata": {
         "id": "mMiGrpYXLFNP"
       },
@@ -358,7 +358,7 @@
     },
     {
       "cell_type": "code",
-      "execution_count": 10,
+      "execution_count": null,
       "metadata": {
         "id": "NfjxfLlujLZi"
       },
@@ -2380,12 +2380,21 @@
     {
       "cell_type": "markdown",
       "source": [
-        "#### 3.6 Data Analytics: RAF-DB"
+        "### 3.6 Data Analytics: RAF-DB"
       ],
       "metadata": {
         "id": "ZDBX7GJ_IpVi"
       }
     },
+    {
+      "cell_type": "markdown",
+      "source": [
+        "Real-world Affective Faces Database (RAF-DB) is a large-scale facial expression database with around 30.000 great-diverse facial images downloaded from the Internet. Based on the crowdsourcing annotation, each image has been independently labeled by about 40 annotators. Images in this database are of the great variability in subjects' age, gender and ethnicity, head poses, lighting conditions, occlusions like glasses, facial hair, or self-occlusion). The dataset is not available for students, so we use an open-source version of 12.000 samples. Hence, the RAF-DB analysis will concentrate only on this subset. We put an additional dataset for only the testing phase. This choice is due to the complexity of the merged dataset, which includes high variability and expression complexity on sample faces, especially for the AffectNet. Hence, we use an aligned and easy dataset called RAF-DB to test the result models on it and compare results with the testing set generated by the final merged dataset. "
+      ],
+      "metadata": {
+        "id": "Xz1WwYviH2AG"
+      }
+    },
     {
       "cell_type": "code",
       "source": [
@@ -2415,7 +2424,7 @@
           "base_uri": "https://localhost:8080/"
         }
       },
-      "execution_count": 24,
+      "execution_count": null,
       "outputs": [
         {
           "output_type": "stream",
@@ -2453,7 +2462,7 @@
           "height": 352
         }
       },
-      "execution_count": 22,
+      "execution_count": null,
       "outputs": [
         {
           "output_type": "display_data",
@@ -2503,7 +2512,7 @@
           "height": 646
         }
       },
-      "execution_count": 26,
+      "execution_count": null,
       "outputs": [
         {
           "output_type": "stream",
@@ -2615,7 +2624,7 @@
           "height": 339
         }
       },
-      "execution_count": 34,
+      "execution_count": null,
       "outputs": [
         {
           "output_type": "execute_result",
@@ -2660,7 +2669,7 @@
           "base_uri": "https://localhost:8080/"
         }
       },
-      "execution_count": 29,
+      "execution_count": null,
       "outputs": [
         {
           "output_type": "stream",
@@ -2693,7 +2702,7 @@
           "height": 297
         }
       },
-      "execution_count": 30,
+      "execution_count": null,
       "outputs": [
         {
           "output_type": "display_data",
@@ -2732,7 +2741,7 @@
           "base_uri": "https://localhost:8080/"
         }
       },
-      "execution_count": 32,
+      "execution_count": null,
       "outputs": [
         {
           "output_type": "stream",
@@ -2761,7 +2770,7 @@
           "height": 339
         }
       },
-      "execution_count": 35,
+      "execution_count": null,
       "outputs": [
         {
           "output_type": "execute_result",
@@ -2801,7 +2810,7 @@
           "base_uri": "https://localhost:8080/"
         }
       },
-      "execution_count": 36,
+      "execution_count": null,
       "outputs": [
         {
           "output_type": "stream",
@@ -3469,7 +3478,7 @@
           "height": 336
         }
       },
-      "execution_count": 39,
+      "execution_count": null,
       "outputs": [
         {
           "output_type": "display_data",