Update app.py

app.py

@@ -1,83 +1,35 @@
-# app.py  ─────────────────────────────────────────────────────────────
-import os
 import gradio as gr
-import spaces                        # ← tells HF/ZeroGPU we need a card
+import spaces
+from detection import EyesDetection                 # comes from imgutils
 
-from detection import (              # your existing helper classes
-    EyesDetection, FaceDetection, HeadDetection, PersonDetection,
-    HandDetection, CensorDetection, HalfBodyDetection,
-    NudeNetDetection, BooruYOLODetection,
-)
-
-# ---------------------------------------------------------------------
-# 0. Global style (unchanged)
-_GLOBAL_CSS = """
-.limit-height { max-height: 55vh; }
-"""
-
-# ---------------------------------------------------------------------
-# 1. Thin, GPU-decorated function that Unity will call
-@spaces.GPU()                        # allocate GPU only while this runs
-def eyes_detect_api(image):
+# -------------------------------------------------------------------
+# 1) GPU-decorated inference function
+@spaces.GPU(estimate=60)            # ask ZeroGPU for up to 60 s of GPU
+def eye_detect(image):
     """
     Args
     ----
-    image : PIL.Image (gradio passes this automatically)
+    image : PIL.Image (gradio supplies this)
 
     Returns
     -------
-    masks : list[PIL.Image]  – one mask per eye
-    boxes : list[list[int]]  – [x1, y1, x2, y2] per eye
+    masks : list[PIL.Image]         – per-eye mask PNGs
+    boxes : list[list[int]]         – [x1, y1, x2, y2] per eye
     """
-    # — lazy-load & cache the model so we don't pay the cost each call
-    global _EYE_DETECTOR
-    if "_EYE_DETECTOR" not in globals():
-        _EYE_DETECTOR = EyesDetection()           # heavy init happens once
-    masks, boxes = _EYE_DETECTOR.process_image(image)
+    global _DET
+    if "_DET" not in globals():     # lazy-load & cache model weights
+        _DET = EyesDetection()
+    masks, boxes = _DET.process_image(image)
     return masks, boxes
 
-# ---------------------------------------------------------------------
-# 2. Build the original UI
-with gr.Blocks(css=_GLOBAL_CSS) as demo:
-    with gr.Row():
-        with gr.Column():
-            gr.HTML('<h2 style="text-align: center;">Object Detections For Anime</h2>')
-            gr.Markdown(
-                'This is the online demo for detection functions of '
-                '[imgutils.detect](https://dghs-imgutils.deepghs.org/main/api_doc/detect/index.html). '
-                'You can try them yourselves with `pip install dghs-imgutils`.'
-            )
-
-    with gr.Row():
-        with gr.Tabs():
-            with gr.Tab('Face Detection'):
-                FaceDetection().make_ui()
-            with gr.Tab('Head Detection'):
-                HeadDetection().make_ui()
-            with gr.Tab('Person Detection'):
-                PersonDetection().make_ui()
-            with gr.Tab('Half Body Detection'):
-                HalfBodyDetection().make_ui()
-            with gr.Tab('Eyes Detection'):
-                EyesDetection().make_ui()
-            with gr.Tab('Hand Detection'):
-                HandDetection().make_ui()
-            with gr.Tab('Censor Point Detection'):
-                CensorDetection().make_ui()
-            with gr.Tab('NudeNet'):
-                NudeNetDetection().make_ui()
-            with gr.Tab('BooruYOLO'):
-                BooruYOLODetection().make_ui()
-
-    # -----------------------------------------------------------------
-    # 3. API-only endpoint (no UI) for Unity / curl
+# -------------------------------------------------------------------
+# 2) API-only Gradio app (no GUI at all)
+with gr.Blocks() as demo:
     demo.api(
-        fn=eyes_detect_api,
+        fn=eye_detect,
         inputs=gr.Image(type="pil"),
         outputs=[gr.Gallery(label="Masks"), gr.JSON(label="Boxes")],
         name="eye_detect"
     )
 
-# ---------------------------------------------------------------------
-# 4. Queue & launch (unchanged except we let Gradio choose CPU threads)
-demo.queue(concurrency_count=os.cpu_count()).launch()
+demo.queue(concurrency_count=1).launch()