add models

app.py

@@ -204,6 +204,8 @@ def ncut_run(
     video_output=False,
 ):
     logging_str = ""
+    resolution = RES_DICT[model_name]
+    logging_str += f"Resolution: {resolution}\n"
     if perplexity >= num_sample_tsne or n_neighbors >= num_sample_tsne:
         # raise gr.Error("Perplexity must be less than the number of samples for t-SNE.")
         gr.Warning("Perplexity/n_neighbors must be less than the number of samples.\n" f"Setting Perplexity to {num_sample_tsne-1}.")
@@ -585,9 +587,10 @@ def make_output_images_section():
 
 def make_parameters_section():
     gr.Markdown("### Parameters <a style='color: #0044CC;' href='https://ncut-pytorch.readthedocs.io/en/latest/how_to_get_better_segmentation/' target='_blank'>Help</a>")
-    from ncut_pytorch.backbone import get_demo_model_names
-    model_names = get_demo_model_names()
-    model_dropdown = gr.Dropdown(model_names, label="Backbone", value="DiNO(dino_vitb8)", elem_id="model_name")
+    from ncut_pytorch.backbone import list_models, get_demo_model_names
+    model_names = list_models()
+    model_names = sorted(model_names)
+    model_dropdown = gr.Dropdown(model_names, label="Backbone", value="DiNO(dino_vitb8_448)", elem_id="model_name")
     layer_slider = gr.Slider(1, 12, step=1, label="Backbone: Layer index", value=10, elem_id="layer")
     node_type_dropdown = gr.Dropdown(["attn: attention output", "mlp: mlp output", "block: sum of residual"], label="Backbone: Layer type", value="block: sum of residual", elem_id="node_type", info="which feature to take from each layer?")
     num_eig_slider = gr.Slider(1, 1000, step=1, label="NCUT: Number of eigenvectors", value=100, elem_id="num_eig", info='increase for more clusters')

requirements.txt

@@ -1,6 +1,6 @@
 torch
 torchvision 
-ncut-pytorch>=1.2.0
+ncut-pytorch>=1.2.6
 opencv-python
 decord
 transformers