diff --git a/python/examples/pipeline/imagenet/benchmark.py b/python/examples/pipeline/imagenet/benchmark.py
index f96d25350fb155140ea5279c802cb7b929e5a12f..22c1ce873f2db2d866a96ff9a68ffce3bd5b98f5 100644
--- a/python/examples/pipeline/imagenet/benchmark.py
+++ b/python/examples/pipeline/imagenet/benchmark.py
@@ -46,11 +46,16 @@ def run_http(idx, batch_size):
     with open(os.path.join(".", "daisy.jpg"), 'rb') as file:
         image_data1 = file.read()
     image = cv2_to_base64(image_data1)
-    data = {"key": ["image"], "value": [image]}
+    keys, values = [], []
+    for i in range(batch_size):
+        keys.append("image_{}".format(i))
+        values.append(image)
+    data = {"key": keys, "value": values}
     start_time = time.time()
     while True:
         r = requests.post(url=url, data=json.dumps(data))
-        if time.time() - start_time > 10:
+        print(r.json())
+        if time.time() - start_time > 20:
             break
     end = time.time()
     return [[end - start]]
diff --git a/python/examples/pipeline/imagenet/resnet50_web_service.py b/python/examples/pipeline/imagenet/resnet50_web_service.py
index 635636cdd3f139065128c4a013d7999c223d87ec..11fa1ab14181fca1833cbbe6c9bb89afa9531b3f 100644
--- a/python/examples/pipeline/imagenet/resnet50_web_service.py
+++ b/python/examples/pipeline/imagenet/resnet50_web_service.py
@@ -38,15 +38,19 @@ class ImagenetOp(Op):
 
     def preprocess(self, input_dicts, data_id, log_id):
         (_, input_dict), = input_dicts.items()
-        data = base64.b64decode(input_dict["image"].encode('utf8'))
-        data = np.fromstring(data, np.uint8)
-        # Note: class variables(self.var) can only be used in process op mode
-        im = cv2.imdecode(data, cv2.IMREAD_COLOR)
-        img = self.seq(im)
-        return {"image": img[np.newaxis, :].copy()}, False, None, ""
+        batch_size = len(input_dict.keys())
+        imgs = []
+        for key in input_dict.keys():
+            data = base64.b64decode(input_dict[key].encode('utf8'))
+            data = np.fromstring(data, np.uint8)
+            im = cv2.imdecode(data, cv2.IMREAD_COLOR)
+            img = self.seq(im)
+            imgs.append(img[np.newaxis, :].copy())
+        input_imgs = np.concatenate(imgs, axis=0)
+        return {"image": input_imgs}, False, None, ""
 
     def postprocess(self, input_dicts, fetch_dict, log_id):
-        score_list = fetch_dict["score"]
+        score_list = fetch_dict["prediction"]
         result = {"label": [], "prob": []}
         for score in score_list:
             score = score.tolist()