diff --git a/README.md b/README.md
index 7e9f1e0..b203d8e 100644
--- a/README.md
+++ b/README.md
@@ -14,13 +14,13 @@ This Operator generates feature vectors from the pytorch pretrained **Resnet50**
 
 **Args:**
 
-​    model_name (`str`):
+- model_name (`str`):
 
-​        The model name for embedding, for example 'resnet50'.
+  The model name for embedding, for example 'resnet50'.
 
-​    framework (`str`):
+- framework (`str`):
 
-​        The framework of the model, the default is 'pytorch'.
+  The framework of the model, the default is 'pytorch'.
 
 `__call__(self, img_tensor: torch.Tensor)`
 
@@ -32,7 +32,7 @@ This Operator generates feature vectors from the pytorch pretrained **Resnet50**
 
 **Returns:**
 
-​    (`Tuple[('cnn', numpy.ndarray)]`)
+​    (`Tuple[('feature_vector', numpy.ndarray)]`)
 
 ​         The embedding of image.
 
@@ -42,7 +42,7 @@ You can get the required python package by [requirements.txt](./requirements.txt
 
 ## How it works
 
-The `towhee/resnet50-image-embedding` Operator implements the function of image embedding, which can add to the pipeline, for example, it's the key Operator named embedding_model within [image_embedding_resnet50](https://hub.towhee.io/towhee/image-embedding-resnet50) pipeline, and it is the red box in the picture below.
+The `towhee/resnet50-image-embedding` Operator implements the function of image embedding, which can add to the pipeline, for example, it's the key Operator named embedding_model within [image-embedding-resnet50](https://hub.towhee.io/towhee/image-embedding-resnet50) pipeline, and it is the red box in the picture below.
 
 ![img](./readme_res/operator.png)
 
diff --git a/resnet50_image_embedding.py b/resnet50_image_embedding.py
index 8e9a7cf..a3401fd 100644
--- a/resnet50_image_embedding.py
+++ b/resnet50_image_embedding.py
@@ -33,7 +33,7 @@ class Resnet50ImageEmbedding(Operator):
             from pytorch.model import Model
         self.model = Model(model_name)
 
-    def __call__(self, img_tensor: torch.Tensor) -> NamedTuple('Outputs', [('cnn', numpy.ndarray)]):
+    def __call__(self, img_tensor: torch.Tensor) -> NamedTuple('Outputs', [('feature_vector', numpy.ndarray)]):
         embedding = self.model(img_tensor)
-        Outputs = NamedTuple('Outputs', [('cnn', numpy.ndarray)])
+        Outputs = NamedTuple('Outputs', [('feature_vector', numpy.ndarray)])
         return Outputs(embedding)
diff --git a/resnet50_image_embedding.yaml b/resnet50_image_embedding.yaml
index 7d1a2ba..ec63089 100644
--- a/resnet50_image_embedding.yaml
+++ b/resnet50_image_embedding.yaml
@@ -10,4 +10,4 @@ call:
   input:
     img_tensor: torch.Tensor
   output:
-    cnn: numpy.ndarray
+    feature_vector: numpy.ndarray