Update app.py

app.py

@@ -12,17 +12,20 @@ import spaces
 ver=[0,0,0,0,0,0,6,7,8,9,10,11]
 ltr=["n","s","m","1","x"]
 tsk=["","-seg","-pose","-obb","-cls"]
-#yolov8s.pt
-modin=f"yolov{ver[9]}{ltr[1]}{tsk[0]}.pt"
-model = YOLO(modin)
 annotators = ["Box","RoundBox","BoxCorner","Color",
               "Circle","Dot","Triangle","Elipse","Halo",
               "PercentageBar","Mask","Polygon","Label",
               "RichLabel","Icon","Crop","Blur","Pixelate","HeatMap"]
+def model_select(v,l,t):
+    modin=f"yolov{v}{l}{t}.pt"
+    print(modin)
+    global model
+    model = YOLO(modin)
+
 
 @spaces.GPU
-def stream_object_detection(video):
-    SUBSAMPLE=1
+def stream_object_detection(video,anno):
+    SUBSAMPLE=2
     cap = cv2.VideoCapture(video)
     # This means we will output mp4 videos
     video_codec = cv2.VideoWriter_fourcc(*"mp4v") # type: ignore
@@ -40,9 +43,10 @@ def stream_object_detection(video):
         frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
         result = model(Image.fromarray(frame))[0]
         detections = sv.Detections.from_ultralytics(result)
-        print(detections)
+        #print(detections)
 
-        box_annotator = eval(f'sv.{annotators[0]}Annotator()')
+        box_annotator = eval(f'sv.{anno}Annotator()')
+        #box_annotator = eval(f'sv.{annotators[0]}Annotator()')
         
         outp = box_annotator.annotate(
           scene=frame.copy(),
@@ -60,16 +64,27 @@ def stream_object_detection(video):
         output_video = cv2.VideoWriter(output_video_name, video_codec, desired_fps, (width, height)) # type: ignore
         iterating, frame = cap.read()
         n_frames += 1
-
-with gr.Blocks() as app:
-    gr.HTML("<div style='font-size: 50px;font-weight: 800;'>Supervision</div><div style='font-size: 30px;'>Video Object Detection</div><div>Github:<a href='https://github.com/roboflow/supervision' target='_blank'>https://github.com/roboflow/supervision</a></div>")
+#css="body{background:aliceblue;}"
+with gr.Blocks(theme="Nymbo/Nymbo_Theme_5") as app:
+    gr.HTML("<div style='font-size: 50px;font-weight: 800;'>SuperVision</div><div style='font-size: 30px;'>Video Object Detection</div><div>Github:<a href='https://github.com/roboflow/supervision' target='_blank'>https://github.com/roboflow/supervision</a></div>")
     #inp = gr.Image(type="filepath")
     with gr.Row():
         with gr.Column():
-            inp = gr.Video()
+            inp = gr.Video(height=300)
             btn = gr.Button()
-        outp_v = gr.Video(label="Processed Video", streaming=True, autoplay=True)
-    outp_j = gr.JSON()
+            with gr.Accordion("Controls",open=False):
+                with gr.Group():
+                    dd1=gr.Dropdown(label="Version",choices=ver[6:],value=ver[9],allow_custom_value=True)
+                    dd2=gr.Dropdown(label="Ltr", choices=ltr,value=ltr[1],allow_custom_value=True)
+                    dd3=gr.Dropdown(label="Task",choices=tsk,value=tsk[0],allow_custom_value=True)
+                    dd4=gr.Dropdown(label="Annotator",choices=annotators,value="Box")
+        with gr.Column():
+            outp_v = gr.Video(label="Processed Video", streaming=True, autoplay=True,height=300)
+            outp_j = gr.JSON()
 
-    btn.click(stream_object_detection,inp,[outp_v,outp_j])
-app.queue().launch()
+    btn.click(stream_object_detection,[inp,dd4],[outp_v,outp_j])
+    app.load(model_select,[dd1,dd2,dd3],None)
+    dd1.change(model_select,[dd1,dd2,dd3],None)    
+    dd2.change(model_select,[dd1,dd2,dd3],None)    
+    dd3.change(model_select,[dd1,dd2,dd3],None)    
+app.queue().launch()