diff --git a/argv.py b/argv.py
index 6725cb7..f518edf 100644
--- a/argv.py
+++ b/argv.py
@@ -70,7 +70,7 @@ class ArgvParser:
                 _, extension = os.path.splitext(a.input)
                 a.output = "output{}".format(extension)
             elif a.output and os.path.isfile(a.input) and os.path.splitext(a.output)[1] \
-                    not in cv2_supported_extension():
+                    not in cv2_supported_extension() + [".gif"]:
                 ArgvParser.parser.error("Output {} file not a supported format.".format(a.output))
 
         def config_args_altered(a):
@@ -241,6 +241,12 @@ class ArgvParser:
                  "(<x_top_left>,<y_top_left>:<x_bot_right>,<x_bot_right>) and overlay the result on the original image."
         )
 
+        scale_mod.add_argument(
+            "--ignore-size",
+            action="store_true",
+            help="Ignore image size checks."
+        )
+
         def check_steps_args():
             def type_func(a):
                 if not re.match(r"^[0-5]:[0-5]$", a):
diff --git a/main.py b/main.py
index af5a129..fc09888 100644
--- a/main.py
+++ b/main.py
@@ -76,8 +76,6 @@ def select_phases():
         phases = add_tail(phases, ImageToResizedCrop)
     elif conf.args['auto_rescale']:
         phases = add_tail(phases, ImageToRescale)
-    elif not os.path.isfile(conf.args['input']):
-        check_shape(read_image(conf.args['input']))
     return phases
 
 
diff --git a/processing/__init__.py b/processing/__init__.py
index d523a5d..3918874 100644
--- a/processing/__init__.py
+++ b/processing/__init__.py
@@ -13,7 +13,7 @@ import imageio
 
 import argv
 from config import Config as conf
-from utils import camel_case_to_str, cv2_supported_extension, read_image, write_image, json_to_argv
+from utils import camel_case_to_str, cv2_supported_extension, read_image, write_image, json_to_argv, check_shape
 
 
 class Process:
@@ -127,6 +127,11 @@ class ImageTransform(Process):
         self.__starting_step = self._args['steps'][0] if self._args['steps'] else 0
         self.__ending_step = self._args['steps'][1] if self._args['steps'] else None
 
+        if not conf.args['ignore_size']:
+            check_shape(read_image(input_path))
+        else:
+            conf.log.warn('Image Size Requirements Unchecked.')
+
         conf.log.debug("All Phases : {}".format(self.__phases))
         conf.log.debug("To Be Executed Phases : {}".format(self.__phases[self.__starting_step:self.__ending_step]))
 
diff --git a/transform/gan/__init__.py b/transform/gan/__init__.py
index 9590c85..19598db 100644
--- a/transform/gan/__init__.py
+++ b/transform/gan/__init__.py
@@ -10,7 +10,6 @@ from torchvision import transforms as transforms
 
 from config import Config as conf
 from transform import ImageTransform
-from utils import check_shape
 
 
 class ImageTransformGAN(ImageTransform):
@@ -37,8 +36,6 @@ class ImageTransformGAN(ImageTransform):
         :param image: <RGB> image to be transform
         :return: None
         """
-        check_shape(image)
-
         if self.__gpu_ids:
             conf.log.debug("GAN Processing Using GPU IDs: {}".format(self.__gpu_ids))
         else: