Update

README.md

@@ -4,7 +4,7 @@ emoji: 🐢
 colorFrom: yellow
 colorTo: indigo
 sdk: gradio
-sdk_version: 3.19.1
+sdk_version: 3.34.0
 app_file: app.py
 pinned: false
 ---

app.py

@@ -22,12 +22,7 @@ sys.path.insert(0, 'anime_face_landmark_detection')
 
 from CFA import CFA
 
-TITLE = 'kanosawa/anime_face_landmark_detection'
-DESCRIPTION = 'This is an unofficial demo for https://github.com/kanosawa/anime_face_landmark_detection.'
-
-HF_TOKEN = os.getenv('HF_TOKEN')
-MODEL_REPO = 'hysts/anime_face_landmark_detection'
-MODEL_FILENAME = 'checkpoint_landmark_191116.pth'
+DESCRIPTION = '# [kanosawa/anime_face_landmark_detection](https://github.com/kanosawa/anime_face_landmark_detection)'
 
 NUM_LANDMARK = 24
 CROP_SIZE = 128
@@ -39,8 +34,7 @@ def load_sample_image_paths() -> list[pathlib.Path]:
         dataset_repo = 'hysts/sample-images-TADNE'
         path = huggingface_hub.hf_hub_download(dataset_repo,
                                                'images.tar.gz',
-                                               repo_type='dataset',
-                                               use_auth_token=HF_TOKEN)
+                                               repo_type='dataset')
         with tarfile.open(path) as f:
             f.extractall()
     return sorted(image_dir.glob('*'))
@@ -55,9 +49,9 @@ def load_face_detector() -> cv2.CascadeClassifier:
 
 
 def load_landmark_detector(device: torch.device) -> torch.nn.Module:
-    path = huggingface_hub.hf_hub_download(MODEL_REPO,
-                                           MODEL_FILENAME,
-                                           use_auth_token=HF_TOKEN)
+    path = huggingface_hub.hf_hub_download(
+        'public-data/anime_face_landmark_detection',
+        'checkpoint_landmark_191116.pth')
     model = CFA(output_channel_num=NUM_LANDMARK + 1, checkpoint_name=path)
     model.to(device)
     model.eval()
@@ -122,17 +116,25 @@ transform = T.Compose([
     T.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5]),
 ])
 
-func = functools.partial(detect,
-                         face_detector=face_detector,
-                         device=device,
-                         transform=transform,
-                         landmark_detector=landmark_detector)
-
-gr.Interface(
-    fn=func,
-    inputs=gr.Image(label='Input', type='filepath'),
-    outputs=gr.Image(label='Output', type='numpy'),
-    examples=examples,
-    title=TITLE,
-    description=DESCRIPTION,
-).queue().launch(show_api=False)
+fn = functools.partial(detect,
+                       face_detector=face_detector,
+                       device=device,
+                       transform=transform,
+                       landmark_detector=landmark_detector)
+
+with gr.Blocks(css='style.css') as demo:
+    gr.Markdown(DESCRIPTION)
+    with gr.Row():
+        with gr.Column():
+            image = gr.Image(label='Input', type='filepath')
+            run_button = gr.Button('Run')
+        with gr.Column():
+            result = gr.Image(label='Result')
+
+    gr.Examples(examples=examples,
+                inputs=image,
+                outputs=result,
+                fn=fn,
+                cache_examples=os.getenv('CACHE_EXAMPLES') == '1')
+    run_button.click(fn=fn, inputs=image, outputs=result, api_name='predict')
+demo.queue(max_size=15).launch()

requirements.txt

@@ -1,3 +1,3 @@
-opencv-python-headless>=4.5.5.62
-torch>=1.10.1
-torchvision>=0.11.2
+opencv-python-headless>=4.7.0.72
+torch==2.0.1
+torchvision==0.15.2

style.css

@@ -0,0 +1,3 @@
+h1 {
+  text-align: center;
+}