[Bugfix][Multimodal] Parse decoded video frame lists as a single video (#55326)

Signed-off-by: waizuichougou <2082431897@qq.com>
Signed-off-by: waizuichougou <105572299+waizuichougou@users.noreply.github.com>
Co-authored-by: Isotr0py <2037008807@qq.com>
Co-authored-by: Isotr0py <Isotr0py@outlook.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

tests/multimodal/test_parse.py

@@ -56,6 +56,23 @@ def test_frame_size_hwc_chw(frame):
     assert items.get_frame_size(0) == (W, H)
 
 
+@pytest.mark.parametrize(
+    "frames",
+    [
+        [np.zeros((H, W, 3), dtype=np.uint8) for _ in range(2)],
+        [torch.zeros((H, W, 3), dtype=torch.uint8) for _ in range(2)],
+    ],
+)
+def test_parse_video_frame_list_as_single_video(frames):
+    """A list of decoded frames must represent one video item."""
+    items = MultiModalDataParser().parse_mm_data({"video": frames})["video"]
+
+    assert items.get_count() == 1
+    video = items.get(0)
+    assert isinstance(video, np.ndarray)
+    np.testing.assert_array_equal(video, np.stack([np.asarray(f) for f in frames]))
+
+
 @pytest.mark.parametrize(
     "modality,processor_cls",
     [

vllm/multimodal/parse.py

@@ -751,8 +751,14 @@ class MultiModalDataParser:
             return VideoEmbeddingItems(data, self.expected_hidden_size)
 
         data_items: list[VideoItem]
-        if (is_list_of(data, PILImage.Image) and len(data) > 0) or (
-            isinstance(data, (np.ndarray, torch.Tensor)) and data.ndim == 4
+        if (
+            (is_list_of(data, PILImage.Image) and len(data) > 0)
+            or (
+                is_list_of(data, (np.ndarray, torch.Tensor), check="all")
+                and len(data) > 0
+                and all(item.ndim == 3 for item in data)
+            )
+            or (isinstance(data, (np.ndarray, torch.Tensor)) and data.ndim == 4)
         ):
             data_items = [data]
         elif isinstance(data, (np.ndarray, torch.Tensor)):