Upload Subtask3_Outdoor (145 files)

Subtask3_Outdoor/group0300/input_meta.json

@@ -0,0 +1,29 @@
+{
+  "question": "Question:\n\nGiven 4 temporally ordered observation frames from an exocentric action video, which image shows the subsequent visual state following the observation sequence?\n\nThe four images above are temporally ordered observations from an exocentric camera in an outdoor environment. Options A–D show candidate visual states sampled from the same action trajectory observed from the same exocentric perspective.\n\nObserve how the action progresses across the observation frames—track hands, people, objects, and scene context—then select the candidate that depicts the subsequent state following the observation sequence.\n\nOnly output one letter: A, B, C, or D.",
+  "question_dir": "question",
+  "option_dir": "options",
+  "question_images": [
+    "1.jpg",
+    "2.jpg",
+    "3.jpg",
+    "4.jpg"
+  ],
+  "options": [
+    {
+      "label": "A",
+      "image": "A.jpg"
+    },
+    {
+      "label": "B",
+      "image": "B.jpg"
+    },
+    {
+      "label": "C",
+      "image": "C.jpg"
+    },
+    {
+      "label": "D",
+      "image": "D.jpg"
+    }
+  ]
+}

Subtask3_Outdoor/group0301/input_meta.json

@@ -0,0 +1,29 @@
+{
+  "question": "Question:\n\nGiven 4 temporally ordered observation frames from an exocentric action video, which image shows the subsequent visual state following the observation sequence?\n\nThe four images above are temporally ordered observations from an exocentric camera in an outdoor environment. Options A–D show candidate visual states sampled from the same action trajectory observed from the same exocentric perspective.\n\nObserve how the action progresses across the observation frames—track hands, people, objects, and scene context—then select the candidate that depicts the subsequent state following the observation sequence.\n\nOnly output one letter: A, B, C, or D.",
+  "question_dir": "question",
+  "option_dir": "options",
+  "question_images": [
+    "1.jpg",
+    "2.jpg",
+    "3.jpg",
+    "4.jpg"
+  ],
+  "options": [
+    {
+      "label": "A",
+      "image": "A.jpg"
+    },
+    {
+      "label": "B",
+      "image": "B.jpg"
+    },
+    {
+      "label": "C",
+      "image": "C.jpg"
+    },
+    {
+      "label": "D",
+      "image": "D.jpg"
+    }
+  ]
+}

Subtask3_Outdoor/group0302/input_meta.json

@@ -0,0 +1,29 @@
+{
+  "question": "Question:\n\nGiven 4 temporally ordered observation frames from an exocentric action video, which image shows the subsequent visual state following the observation sequence?\n\nThe four images above are temporally ordered observations from an exocentric camera in an outdoor environment. Options A–D show candidate visual states sampled from the same action trajectory observed from the same exocentric perspective.\n\nObserve how the action progresses across the observation frames—track hands, people, objects, and scene context—then select the candidate that depicts the subsequent state following the observation sequence.\n\nOnly output one letter: A, B, C, or D.",
+  "question_dir": "question",
+  "option_dir": "options",
+  "question_images": [
+    "1.jpg",
+    "2.jpg",
+    "3.jpg",
+    "4.jpg"
+  ],
+  "options": [
+    {
+      "label": "A",
+      "image": "A.jpg"
+    },
+    {
+      "label": "B",
+      "image": "B.jpg"
+    },
+    {
+      "label": "C",
+      "image": "C.jpg"
+    },
+    {
+      "label": "D",
+      "image": "D.jpg"
+    }
+  ]
+}

Subtask3_Outdoor/group0303/input_meta.json

@@ -0,0 +1,29 @@
+{
+  "question": "Question:\n\nGiven 4 temporally ordered observation frames from an exocentric action video, which image shows the subsequent visual state following the observation sequence?\n\nThe four images above are temporally ordered observations from an exocentric camera in an outdoor environment. Options A–D show candidate visual states sampled from the same action trajectory observed from the same exocentric perspective.\n\nObserve how the action progresses across the observation frames—track hands, people, objects, and scene context—then select the candidate that depicts the subsequent state following the observation sequence.\n\nOnly output one letter: A, B, C, or D.",
+  "question_dir": "question",
+  "option_dir": "options",
+  "question_images": [
+    "1.jpg",
+    "2.jpg",
+    "3.jpg",
+    "4.jpg"
+  ],
+  "options": [
+    {
+      "label": "A",
+      "image": "A.jpg"
+    },
+    {
+      "label": "B",
+      "image": "B.jpg"
+    },
+    {
+      "label": "C",
+      "image": "C.jpg"
+    },
+    {
+      "label": "D",
+      "image": "D.jpg"
+    }
+  ]
+}

Subtask3_Outdoor/group0304/input_meta.json

@@ -0,0 +1,29 @@
+{
+  "question": "Question:\n\nGiven 4 temporally ordered observation frames from an exocentric action video, which image shows the subsequent visual state following the observation sequence?\n\nThe four images above are temporally ordered observations from an exocentric camera in an outdoor environment. Options A–D show candidate visual states sampled from the same action trajectory observed from the same exocentric perspective.\n\nObserve how the action progresses across the observation frames—track hands, people, objects, and scene context—then select the candidate that depicts the subsequent state following the observation sequence.\n\nOnly output one letter: A, B, C, or D.",
+  "question_dir": "question",
+  "option_dir": "options",
+  "question_images": [
+    "1.jpg",
+    "2.jpg",
+    "3.jpg",
+    "4.jpg"
+  ],
+  "options": [
+    {
+      "label": "A",
+      "image": "A.jpg"
+    },
+    {
+      "label": "B",
+      "image": "B.jpg"
+    },
+    {
+      "label": "C",
+      "image": "C.jpg"
+    },
+    {
+      "label": "D",
+      "image": "D.jpg"
+    }
+  ]
+}

Subtask3_Outdoor/group0305/input_meta.json

@@ -0,0 +1,29 @@
+{
+  "question": "Question:\n\nGiven 4 temporally ordered observation frames from an exocentric action video, which image shows the subsequent visual state following the observation sequence?\n\nThe four images above are temporally ordered observations from an exocentric camera in an outdoor environment. Options A–D show candidate visual states sampled from the same action trajectory observed from the same exocentric perspective.\n\nObserve how the action progresses across the observation frames—track hands, people, objects, and scene context—then select the candidate that depicts the subsequent state following the observation sequence.\n\nOnly output one letter: A, B, C, or D.",
+  "question_dir": "question",
+  "option_dir": "options",
+  "question_images": [
+    "1.jpg",
+    "2.jpg",
+    "3.jpg",
+    "4.jpg"
+  ],
+  "options": [
+    {
+      "label": "A",
+      "image": "A.jpg"
+    },
+    {
+      "label": "B",
+      "image": "B.jpg"
+    },
+    {
+      "label": "C",
+      "image": "C.jpg"
+    },
+    {
+      "label": "D",
+      "image": "D.jpg"
+    }
+  ]
+}