diff --git a/README.md b/README.md
index dc22870..f459b25 100644
--- a/README.md
+++ b/README.md
@@ -17,15 +17,27 @@ This operator extracts features for video or text with [MDMMT: Multidomain Multi
 
 ## Code Example
 
-Load a video embeddings extracted from different upstream expert networks, such as video, RGB, audio.
-
 Read the text to generate a text embedding. 
 
- *Write the pipeline code*:
+```python
+from towhee.dc2 import pipe, ops, DataCollection
+
+p = (
+    pipe.input('text') \
+        .map('text', 'vec', ops.video_text_embedding.mdmmt(modality='text', device='cuda:0')) \
+        .output('text', 'vec')
+)
+
+DataCollection(p('Hello world.')).show()
+```
+![](text_emb_result.png)
+
+Load a video embeddings extracted from different upstream expert networks, such as video, RGB, audio.
 
 ```python
-import towhee
 import torch
+from towhee.dc2 import pipe, ops, DataCollection
+
 
 torch.manual_seed(42)
 
@@ -52,14 +64,16 @@ features_ind = {
 
 video_input_dict = {"features": features, "features_t": features_t, "features_ind": features_ind}
 
-towhee.dc([video_input_dict]).video_text_embedding.mdmmt(modality='video', device='cpu').show()
+p = (
+    pipe.input('video_input_dict') \
+        .map('video_input_dict', 'vec', ops.video_text_embedding.mdmmt(modality='video', device='cuda:0')) \
+        .output('video_input_dict', 'vec')
+)
 
-towhee.dc(['Hello world.']).video_text_embedding.mdmmt(modality='text', device='cpu').show()
+DataCollection(p(video_input_dict)).show()
 ```
-![](vect_simplified_video.png)
-![](vect_simplified_text.png)
+![](video_emb_result.png)
 
-*Write a same pipeline with explicit inputs/outputs name specifications:*
 
 <br />
 
diff --git a/text_emb_result.png b/text_emb_result.png
new file mode 100644
index 0000000..b7b2a21
Binary files /dev/null and b/text_emb_result.png differ
diff --git a/vect_simplified_text.png b/vect_simplified_text.png
deleted file mode 100644
index 91004b6..0000000
Binary files a/vect_simplified_text.png and /dev/null differ
diff --git a/vect_simplified_video.png b/vect_simplified_video.png
deleted file mode 100644
index 056d876..0000000
Binary files a/vect_simplified_video.png and /dev/null differ
diff --git a/video_emb_result.png b/video_emb_result.png
new file mode 100644
index 0000000..7bc5c68
Binary files /dev/null and b/video_emb_result.png differ