Update app.py
Browse files
app.py
CHANGED
|
@@ -827,7 +827,9 @@ with gr.Blocks(title="Salamandra Vision 7B 路 ZeroGPU", css=custom_css,theme=gr.
|
|
| 827 |
gr.Markdown('<h2 style="text-align:center">Embeddings facials</h2>')
|
| 828 |
with gr.Row():
|
| 829 |
face_img = gr.Image(label="Imatge per embedding facial", type="pil")
|
|
|
|
| 830 |
face_btn = gr.Button("Obt茅 embedding facial", variant="primary")
|
|
|
|
| 831 |
face_out = gr.JSON(label="Embedding facial (vector)")
|
| 832 |
face_btn.click(face_image_embedding, [face_img], face_out, api_name="face_image_embedding", concurrency_limit=1)
|
| 833 |
gr.Markdown("---")
|
|
@@ -841,7 +843,9 @@ with gr.Blocks(title="Salamandra Vision 7B 路 ZeroGPU", css=custom_css,theme=gr.
|
|
| 841 |
threshold = gr.Slider(0.0, 100.0, value=30.0, step=1.0, label="Llindar")
|
| 842 |
offset_frames = gr.Slider(0, 30, value=5, step=1, label="Despla莽ament de frames")
|
| 843 |
crop_ratio = gr.Slider(0.0, 1.0, value=1.0, step=0.05, label="Ra贸 de retall")
|
|
|
|
| 844 |
scenes_btn = gr.Button("Extreu escenes", variant="primary")
|
|
|
|
| 845 |
scenes_gallery_out = gr.Gallery(label="Fotogrames clau de l鈥檈scena", show_label=False, columns=4, height="auto")
|
| 846 |
scenes_info_out = gr.JSON(label="Informaci贸 de l鈥檈scena")
|
| 847 |
|
|
@@ -860,7 +864,9 @@ with gr.Blocks(title="Salamandra Vision 7B 路 ZeroGPU", css=custom_css,theme=gr.
|
|
| 860 |
gr.Markdown('<h2 style="text-align:center">Descripci贸 per lots amb Salamandra Vision</h2>')
|
| 861 |
with gr.Row():
|
| 862 |
img_input = gr.Gallery(label="Llot d鈥檌matges", show_label=False)
|
|
|
|
| 863 |
describe_btn = gr.Button("Genera descripcions", variant="primary")
|
|
|
|
| 864 |
desc_output = gr.Textbox(label="Descripcions de les imatges")
|
| 865 |
|
| 866 |
describe_btn.click(
|
|
@@ -883,12 +889,15 @@ with gr.Blocks(title="Salamandra Vision 7B 路 ZeroGPU", css=custom_css,theme=gr.
|
|
| 883 |
placeholder='{"index": 0, "start": 0.0, "end": 1.2}',
|
| 884 |
lines=3
|
| 885 |
)
|
|
|
|
| 886 |
faces_input = gr.Textbox(
|
| 887 |
label="Llistat de diccionaris face_col (format JSON)",
|
| 888 |
placeholder='[{"nombre": "Anna", "embedding": [0.12, 0.88, ...]}, ...]',
|
| 889 |
lines=5
|
| 890 |
)
|
|
|
|
| 891 |
process_btn = gr.Button("Processar imatge (OCR + Persones)", variant="primary")
|
|
|
|
| 892 |
output_json = gr.JSON(label="Resultat complet")
|
| 893 |
|
| 894 |
process_btn.click(
|
|
|
|
| 827 |
gr.Markdown('<h2 style="text-align:center">Embeddings facials</h2>')
|
| 828 |
with gr.Row():
|
| 829 |
face_img = gr.Image(label="Imatge per embedding facial", type="pil")
|
| 830 |
+
with gr.Row():
|
| 831 |
face_btn = gr.Button("Obt茅 embedding facial", variant="primary")
|
| 832 |
+
with gr.Row():
|
| 833 |
face_out = gr.JSON(label="Embedding facial (vector)")
|
| 834 |
face_btn.click(face_image_embedding, [face_img], face_out, api_name="face_image_embedding", concurrency_limit=1)
|
| 835 |
gr.Markdown("---")
|
|
|
|
| 843 |
threshold = gr.Slider(0.0, 100.0, value=30.0, step=1.0, label="Llindar")
|
| 844 |
offset_frames = gr.Slider(0, 30, value=5, step=1, label="Despla莽ament de frames")
|
| 845 |
crop_ratio = gr.Slider(0.0, 1.0, value=1.0, step=0.05, label="Ra贸 de retall")
|
| 846 |
+
with gr.Row():
|
| 847 |
scenes_btn = gr.Button("Extreu escenes", variant="primary")
|
| 848 |
+
with gr.Row():
|
| 849 |
scenes_gallery_out = gr.Gallery(label="Fotogrames clau de l鈥檈scena", show_label=False, columns=4, height="auto")
|
| 850 |
scenes_info_out = gr.JSON(label="Informaci贸 de l鈥檈scena")
|
| 851 |
|
|
|
|
| 864 |
gr.Markdown('<h2 style="text-align:center">Descripci贸 per lots amb Salamandra Vision</h2>')
|
| 865 |
with gr.Row():
|
| 866 |
img_input = gr.Gallery(label="Llot d鈥檌matges", show_label=False)
|
| 867 |
+
with gr.Row():
|
| 868 |
describe_btn = gr.Button("Genera descripcions", variant="primary")
|
| 869 |
+
with gr.Row():
|
| 870 |
desc_output = gr.Textbox(label="Descripcions de les imatges")
|
| 871 |
|
| 872 |
describe_btn.click(
|
|
|
|
| 889 |
placeholder='{"index": 0, "start": 0.0, "end": 1.2}',
|
| 890 |
lines=3
|
| 891 |
)
|
| 892 |
+
with gr.Row():
|
| 893 |
faces_input = gr.Textbox(
|
| 894 |
label="Llistat de diccionaris face_col (format JSON)",
|
| 895 |
placeholder='[{"nombre": "Anna", "embedding": [0.12, 0.88, ...]}, ...]',
|
| 896 |
lines=5
|
| 897 |
)
|
| 898 |
+
with gr.Row():
|
| 899 |
process_btn = gr.Button("Processar imatge (OCR + Persones)", variant="primary")
|
| 900 |
+
with gr.Row():
|
| 901 |
output_json = gr.JSON(label="Resultat complet")
|
| 902 |
|
| 903 |
process_btn.click(
|