change gray image code from g to y

dataloaders/kitti_loader.py

@@ -12,7 +12,7 @@
 from dataloaders import transforms
 from dataloaders.pose_estimator import get_pose_pnp
 
-input_options = ['d', 'rgb', 'rgbd', 'g', 'gd']
+input_options = ['d', 'rgb', 'rgbd', 'y', 'yd']
 
 def load_calib():
     """
@@ -35,7 +35,7 @@ def load_calib():
 root_d = os.path.join('..', 'data', 'kitti_depth')
 root_rgb = os.path.join('..', 'data', 'kitti_rgb')
 def get_paths_and_transform(split, args):
-    assert (args.use_d or args.use_rgb or args.use_g), 'no proper input selected'
+    assert (args.use_d or args.use_rgb or args.use_y), 'no proper input selected'
 
     if split == "train":
         transform = train_transform
@@ -94,7 +94,7 @@ def get_rgb_paths(p):
         raise(RuntimeError("Requested sparse depth but none was found"))
     if len(paths_rgb) == 0 and args.use_rgb:
         raise(RuntimeError("Requested rgb images but none was found"))
-    if len(paths_rgb) == 0 and args.use_g:
+    if len(paths_rgb) == 0 and args.use_y:
         raise(RuntimeError("Requested gray images but no rgb was found"))
     if len(paths_rgb) != len(paths_d) or len(paths_rgb) != len(paths_gt):
         raise(RuntimeError("Produced different sizes for datasets"))
@@ -186,7 +186,7 @@ def no_transform(rgb, sparse, target, rgb_near, args):
 def handle_gray(rgb, args):
     if rgb is None:
         return None, None
-    if not args.use_g:
+    if not args.use_y:
         return rgb, None
     else:
         img = np.array(Image.fromarray(rgb).convert('L'))
@@ -238,7 +238,7 @@ def __init__(self, split, args):
 
     def __getraw__(self, index):
         rgb = rgb_read(self.paths['rgb'][index]) if \
-            (self.paths['rgb'][index] is not None and (self.args.use_rgb or self.args.use_g)) else None
+            (self.paths['rgb'][index] is not None and (self.args.use_rgb or self.args.use_y)) else None
         sparse = depth_read(self.paths['d'][index]) if \
             (self.paths['d'][index] is not None and self.args.use_d) else None
         target = depth_read(self.paths['gt'][index]) if \
@@ -265,7 +265,7 @@ def __getitem__(self, index):
 
         rgb, gray = handle_gray(rgb, self.args)
         candidates = {"rgb":rgb, "d":sparse, "gt":target, \
-            "g":gray, "r_mat":r_mat, "t_vec":t_vec, "rgb_near":rgb_near}
+            "y":gray, "r_mat":r_mat, "t_vec":t_vec, "rgb_near":rgb_near}
         items = {key:to_float_tensor(val) for key, val in candidates.items() if val is not None}
 
         return items

main.py

@@ -36,7 +36,7 @@
                     metavar='N', help='print frequency (default: 10)')
 parser.add_argument('--resume', default='', type=str, metavar='PATH',
                     help='path to latest checkpoint (default: none)')
-parser.add_argument('-i','--input', type=str, default='gd',
+parser.add_argument('-i','--input', type=str, default='yd',
                     choices=input_options, help='input: | '.join(input_options))
 parser.add_argument('-l','--layers', type=int, default=34,
                     help='use 16 for sparse_conv; use 18 or 34 for resnet')
@@ -60,7 +60,7 @@
 args.result = os.path.join('..', 'results')
 args.use_rgb = ('rgb' in args.input) or args.use_pose
 args.use_d = 'd' in args.input
-args.use_g = 'g' in args.input
+args.use_y = 'y' in args.input
 if args.use_pose:
     args.w1, args.w2 = 0.1, 0.1
 else:

model.py

@@ -65,7 +65,7 @@ def __init__(self, args):
         if 'rgb' in self.modality:
             channels = 64 * 3 // len(self.modality)
             self.conv1_img = conv_bn_relu(3, channels, kernel_size=3, stride=1, padding=1)
-        elif 'g' in self.modality:
+        elif 'y' in self.modality:
             channels = 64 // len(self.modality)
             self.conv1_img = conv_bn_relu(1, channels, kernel_size=3, stride=1, padding=1)
 
@@ -107,10 +107,10 @@ def forward(self, x):
             conv1_d = self.conv1_d(x['d'])
         if 'rgb' in self.modality:
             conv1_img = self.conv1_img(x['rgb'])
-        elif 'g' in self.modality:
-            conv1_img = self.conv1_img(x['g'])
+        elif 'y' in self.modality:
+            conv1_img = self.conv1_img(x['y'])
 
-        if self.modality=='rgbd' or self.modality=='gd':
+        if self.modality=='rgbd' or self.modality=='yd':
             conv1 = torch.cat((conv1_d, conv1_img),1)
         else:
             conv1 = conv1_d if (self.modality=='d') else conv1_img

vis_utils.py

@@ -24,10 +24,10 @@ def preprocess_depth(x):
         rgb = np.squeeze(ele['rgb'][0,...].data.cpu().numpy())
         rgb = np.transpose(rgb, (1, 2, 0))
         img_list.append(rgb)
-    elif 'g' in ele:
-        g = np.squeeze(ele['g'][0,...].data.cpu().numpy())
-        g = np.array(Image.fromarray(g).convert('RGB'))
-        img_list.append(g)
+    elif 'y' in ele:
+        y = np.squeeze(ele['y'][0,...].data.cpu().numpy())
+        y = np.array(Image.fromarray(y).convert('RGB'))
+        img_list.append(y)
     if 'd' in ele:
         img_list.append(preprocess_depth(ele['d'][0,...]))
     img_list.append(preprocess_depth(pred[0,...]))