diff --git a/libraries/transformer/src/androidTest/java/androidx/media3/transformer/AndroidTestUtil.java b/libraries/transformer/src/androidTest/java/androidx/media3/transformer/AndroidTestUtil.java
index f76ca3a504..36f7c4f1c7 100644
--- a/libraries/transformer/src/androidTest/java/androidx/media3/transformer/AndroidTestUtil.java
+++ b/libraries/transformer/src/androidTest/java/androidx/media3/transformer/AndroidTestUtil.java
@@ -815,17 +815,18 @@ public final class AndroidTestUtil {
    *
    * @param context The {@link Context context}.
    * @param testId The test ID.
-   * @param inputFormat The {@link Format format} to decode.
+   * @param inputFormat The {@link Format format} to decode, or the input is not produced by
+   *     MediaCodec, like an image.
    * @param outputFormat The {@link Format format} to encode/mux or {@code null} if the output won't
    *     be encoded or muxed.
    * @throws AssumptionViolatedException If the device does not support the formats. In this case,
    *     the reason for skipping the test is logged.
    */
   public static void assumeFormatsSupported(
-      Context context, String testId, Format inputFormat, @Nullable Format outputFormat)
+      Context context, String testId, @Nullable Format inputFormat, @Nullable Format outputFormat)
       throws IOException, JSONException, MediaCodecUtil.DecoderQueryException {
     // TODO(b/278657595): Make this capability check match the default codec factory selection code.
-    boolean canDecode = canDecode(inputFormat);
+    boolean canDecode = inputFormat == null || canDecode(inputFormat);
 
     boolean canEncode = outputFormat == null || canEncode(outputFormat);
     boolean canMux = outputFormat == null || canMux(outputFormat);
diff --git a/libraries/transformer/src/androidTest/java/androidx/media3/transformer/TransformerMixedInputEndToEndTest.java b/libraries/transformer/src/androidTest/java/androidx/media3/transformer/TransformerMixedInputEndToEndTest.java
index 2c6ba0d108..8679596651 100644
--- a/libraries/transformer/src/androidTest/java/androidx/media3/transformer/TransformerMixedInputEndToEndTest.java
+++ b/libraries/transformer/src/androidTest/java/androidx/media3/transformer/TransformerMixedInputEndToEndTest.java
@@ -41,6 +41,8 @@ import androidx.test.core.app.ApplicationProvider;
 import androidx.test.ext.junit.runners.AndroidJUnit4;
 import com.google.common.collect.ImmutableList;
 import java.io.File;
+import java.util.ArrayList;
+import java.util.List;
 import org.junit.Before;
 import org.junit.Rule;
 import org.junit.Test;
@@ -77,6 +79,8 @@ public class TransformerMixedInputEndToEndTest {
     testId = testName.getMethodName();
   }
 
+  // TODO: b/343362776 - Add tests to assert enough silence is generated.
+
   @Test
   public void videoEditing_withImageThenVideoInputs_completesWithCorrectFrameCount()
       throws Exception {
@@ -133,6 +137,47 @@ public class TransformerMixedInputEndToEndTest {
     assertThat(new File(result.filePath).length()).isGreaterThan(0);
   }
 
+  @Test
+  public void videoEditing_withShortAlternatingImages_completesWithCorrectFrameCount()
+      throws Exception {
+    Transformer transformer =
+        new Transformer.Builder(context)
+            .setEncoderFactory(
+                new DefaultEncoderFactory.Builder(context).setEnableFallback(false).build())
+            .build();
+
+    EditedMediaItem image1 =
+        new EditedMediaItem.Builder(MediaItem.fromUri(PNG_ASSET_URI_STRING))
+            .setDurationUs(100_000)
+            .setFrameRate(30)
+            .build();
+    int image1FrameCount = 3;
+    EditedMediaItem image2 =
+        new EditedMediaItem.Builder(MediaItem.fromUri(JPG_ASSET_URI_STRING))
+            .setDurationUs(200_000)
+            .setFrameRate(30)
+            .build();
+    int image2FrameCount = 6;
+
+    ArrayList<EditedMediaItem> editedMediaItems = new ArrayList<>(100);
+    for (int i = 0; i < 50; i++) {
+      editedMediaItems.add(image1);
+      editedMediaItems.add(image2);
+    }
+
+    ExportTestResult result =
+        new TransformerAndroidTestRunner.Builder(context, transformer)
+            .build()
+            .run(testId, buildComposition(editedMediaItems));
+
+    // TODO: b/346289922 - Check frame count with extractors.
+    assertThat(result.exportResult.videoFrameCount)
+        .isEqualTo(50 * image1FrameCount + 50 * image2FrameCount);
+    // 50 100ms-images and 50 200ms-images
+    assertThat(result.exportResult.durationMs).isEqualTo(14_966);
+    assertThat(new File(result.filePath).length()).isGreaterThan(0);
+  }
+
   @Test
   public void
       videoEditing_withComplexMixedColorSpaceSdrVideoAndImageInputsEndWithVideo_completesWithCorrectFrameCount()
@@ -343,9 +388,8 @@ public class TransformerMixedInputEndToEndTest {
         .build();
   }
 
-  private static Composition buildComposition(
-      EditedMediaItem editedMediaItem, EditedMediaItem... editedMediaItems) {
-    return new Composition.Builder(new EditedMediaItemSequence(editedMediaItem, editedMediaItems))
+  private static Composition buildComposition(List<EditedMediaItem> editedMediaItems) {
+    return new Composition.Builder(new EditedMediaItemSequence(editedMediaItems))
         .setEffects(
             new Effects(
                 /* audioProcessors= */ ImmutableList.of(),
@@ -355,4 +399,13 @@ public class TransformerMixedInputEndToEndTest {
                         /* width= */ 480, /* height= */ 360, Presentation.LAYOUT_SCALE_TO_FIT))))
         .build();
   }
+
+  private static Composition buildComposition(
+      EditedMediaItem editedMediaItem, EditedMediaItem... editedMediaItems) {
+    return buildComposition(
+        new ImmutableList.Builder<EditedMediaItem>()
+            .add(editedMediaItem)
+            .add(editedMediaItems)
+            .build());
+  }
 }