accomodate for new paths

pbnr_pytorch/scene_loaders/ibr_scene.py

@@ -64,7 +64,7 @@ class Camera(nn.Module):
             self.image = nn.Parameter(loaded_optimizedimages.requires_grad_(True))
         else:
             image = image.clamp(0.0, 1.0)
-            self.image = nn.Parameter(image.cuda().requires_grad_(True))
+            self.image = nn.Parameter(image.requires_grad_(True))
 
         self.image_width = self.image.shape[3]
         self.image_height = self.image.shape[2]

pbnr_pytorch/scene_loaders/read_scenes_types.py

@@ -101,7 +101,7 @@ def readColmapSceneInfo_Test(name, path):
         FovY = 2.0 * math.atan(0.5 * height / focal_length_y)
         FovX = 2.0 * math.atan(0.5 * width / focal_length_x)
 
-        image_path = os.path.join(path, "stereo/images", extr.name)
+        image_path = os.path.join(path, "stereo/images", extr.name) + ".png"
         image_name = os.path.basename(image_path).split(".")[0]
         image = Image.open(image_path)
         image = torch.from_numpy(np.array(image)) / 255.0

pbnr_pytorch/test_render_and_save.py

@@ -99,7 +99,8 @@ with torch.no_grad():
     torch.manual_seed(0)
     random.seed(0)
 
-    dataset_name = "crazy_blade2"
+    dataset_name = "hallway_lamp"
+    path_name = "test_path2"
     args.input_path = "/data/graphdeco/user/gkopanas/scenes/catacaustic_new/" + dataset_name + "/sibr/pbnrScene/" + dataset_name + ".json"
 
     with open(args.input_path) as json_file:
@@ -117,7 +118,7 @@ with torch.no_grad():
                            int(args.test_cameras), load_iter)
 
 
-        test_cameras_info = readColmapSceneInfo_Test(dataset_name, r"/data/graphdeco/user/gkopanas/scenes/catacaustic_new/" + dataset_name + "/colmap_1000/test_path_colmap").cameras
+        test_cameras_info = readColmapSceneInfo_Test(dataset_name, r"/data/graphdeco/user/gkopanas/scenes/catacaustic_new/" + dataset_name + "/colmap_1000/" + path_name).cameras
         test_cameras = []
         test_neighbors_dict_path = os.path.join(r"/data/graphdeco/user/gkopanas/scenes/catacaustic_new/" + dataset_name + "/sibr/pbnrScene/" + "neighbors_dict_test.json")
         with open(test_neighbors_dict_path) as json_file:
@@ -145,13 +146,15 @@ with torch.no_grad():
             test_cameras[-1].neighbors = [scene.scenes[0].map_imgfilename_to_idx[image_name] for image_name in test_neighbors_dict[os.path.basename(test_cameras[-1].image_name)]]
 
         print(test_cameras)
-        output = "/data/graphdeco/user/gkopanas/pointbased_neural_rendering/pbnr_pytorch/tensorboard_3d/" + dataset_name + "/path_renders_{}_{}_".format(args.w, args.h) + str(load_iter)
+        output = "/data/graphdeco/user/gkopanas/pointbased_neural_rendering/pbnr_pytorch/tensorboard_3d/" + dataset_name + "/{}_renders_".format(path_name) + str(load_iter)
         makedirs(output, exist_ok=True)
+        makedirs(os.path.join(output, "gt"), exist_ok=True)
+        makedirs(os.path.join(output, "renders"), exist_ok=True)
 
         imgs = torch.tensor([])
         gts = torch.tensor([])
 
-        for view_cam in test_cameras:
+        for idx, view_cam in enumerate(test_cameras):
             print(view_cam.image_name)
             view_list = view_cam.neighbors
             view_cam.image_width = 1000
@@ -161,9 +164,9 @@ with torch.no_grad():
 
             image, image_stack, _ = render_viewpoint(view_cam, pcloud_cams, patch=None)
             torchvision.utils.save_image(image,
-                                         os.path.join(output, "out_{}.png".format(view_cam.image_name)))
+                                         os.path.join(output, "renders", "path_" + str(idx).zfill(4) + ".png"))
             torchvision.utils.save_image(view_cam.image,
-                                         os.path.join(output, "gt_{}.png".format(view_cam.image_name)))
+                                         os.path.join(output, "gt", "path_" + str(idx).zfill(4) + ".png"))
             imgs = torch.cat((imgs, image.cpu()), dim=0)
             gts = torch.cat((gts, view_cam.image.cpu()), dim=0)