diff --git a/README.md b/README.md
index 94dc821..2adf4f4 100644
--- a/README.md
+++ b/README.md
@@ -93,6 +93,18 @@ For training synthetic scenes such as `bouncingballs`, run
 python train.py -s data/dnerf/bouncingballs --port 6017 --expname "dnerf/bouncingballs" --configs arguments/dnerf/bouncingballs.py 
 ```
 
+For training dynerf scenes such as `cut_roasted_beef`, run
+```python
+# First, extract the frames of each video.
+python scripts/preprocess_dynerf.py --datadir data/dynerf/cut_roasted_beef
+# Second, generate point clouds from input data.
+bash colmap.sh data/dynerf/cut_roasted_beef llff
+# Third, downsample the point clouds generated in the second step.
+python scripts/downsample_point.py data/dynerf/cut_roasted_beef/colmap/dense/workspace/fused.ply data/dynerf/cut_roasted_beef/points3D_downsample2.ply
+# Finally, train.
+python train.py -s data/dynerf/cut_roasted_beef --port 6017 --expname "dynerf/cut_roasted_beef" --configs arguments/dynerf/cut_roasted_beef.py 
+```
+
 You can customize your training config through the config files.
 
 Checkpoint
diff --git a/scene/neural_3D_dataset_NDC.py b/scene/neural_3D_dataset_NDC.py
index 7b49877..36a2ded 100644
--- a/scene/neural_3D_dataset_NDC.py
+++ b/scene/neural_3D_dataset_NDC.py
@@ -262,7 +262,7 @@ class Neural3D_NDC_Dataset(Dataset):
         poses_arr = np.load(os.path.join(self.root_dir, "poses_bounds.npy"))
         poses = poses_arr[:, :-2].reshape([-1, 3, 5])  # (N_cams, 3, 5)
         self.near_fars = poses_arr[:, -2:]
-        videos = glob.glob(os.path.join(self.root_dir, "cam*"))
+        videos = glob.glob(os.path.join(self.root_dir, "cam*.mp4"))
         videos = sorted(videos)
         # breakpoint()
         assert len(videos) == poses_arr.shape[0]
diff --git a/scripts/llff2colmap.py b/scripts/llff2colmap.py
index ce13f9b..e836cd5 100644
--- a/scripts/llff2colmap.py
+++ b/scripts/llff2colmap.py
@@ -91,7 +91,7 @@ if not os.path.exists(colmap_dir):
 poses_arr = np.load(os.path.join(root_dir, "poses_bounds.npy"))
 poses = poses_arr[:, :-2].reshape([-1, 3, 5])  # (N_cams, 3, 5)
 near_fars = poses_arr[:, -2:]
-videos = glob.glob(os.path.join(root_dir, "cam*"))
+videos = glob.glob(os.path.join(root_dir, "cam[0-9][0-9]"))
 videos = sorted(videos)
 assert len(videos) == poses_arr.shape[0]
 H, W, focal = poses[0, :, -1]
@@ -109,7 +109,7 @@ poses = np.concatenate([poses[..., 1:2], -poses[..., :1], poses[..., 2:4]], -1)
 # poses[..., 3] /= scale_factor
 # Sample N_views poses for validation - NeRF-like camera trajectory.
 # val_poses = directions
-videos = glob.glob(os.path.join(root_dir, "cam*"))
+videos = glob.glob(os.path.join(root_dir, "cam[0-9][0-9]"))
 videos = sorted(videos)
 image_paths = []
 for index, video_path in enumerate(videos):
diff --git a/scripts/preprocess_dynerf.py b/scripts/preprocess_dynerf.py
new file mode 100644
index 0000000..dc0ed15
--- /dev/null
+++ b/scripts/preprocess_dynerf.py
@@ -0,0 +1,15 @@
+from argparse import ArgumentParser
+import sys
+sys.path.append('./scene')
+from neural_3D_dataset_NDC import Neural3D_NDC_Dataset
+# import scene
+# from scene.neural_3D_dataset_NDC import Neural3D_NDC_Dataset
+
+if __name__ == '__main__':
+    parser = ArgumentParser(description="Extract images from dynerf videos")
+    parser.add_argument("--datadir", default='data/dynerf/cut_roasted_beef', type=str)
+    args = parser.parse_args()
+    train_dataset = Neural3D_NDC_Dataset(args.datadir, "train", 1.0, time_scale=1, 
+                                         scene_bbox_min=[-2.5, -2.0, -1.0], scene_bbox_max=[2.5, 2.0, 1.0], eval_index=0)    
+    test_dataset = Neural3D_NDC_Dataset(args.datadir, "test", 1.0, time_scale=1, 
+                                        scene_bbox_min=[-2.5, -2.0, -1.0], scene_bbox_max=[2.5, 2.0, 1.0], eval_index=0)