fixed SD image size

app/bot.py

@@ -50,6 +50,9 @@ class DailyVision(EventHandler):
 
         self.logger.info(f"Expiration timer set to: {self.__expiration}")
 
+         # Setup camera
+        self.setup_camera()
+
     def run(self, meeting_url, token):
         # Join
         self.logger.info(f"Connecting to room {meeting_url} as {self.__bot_name}")
@@ -81,12 +84,12 @@ class DailyVision(EventHandler):
         # Say hello
         self.wave()
 
-    def setup_camera(self, video_frame):
+    def setup_camera(self):
         if not self.__camera:
             self.__camera = Daily.create_camera_device("camera",
                                                        width = 640,
                                                        height = 480,
-                                                       color_format="RGBA")
+                                                       color_format="RGB")
             self.__client.update_inputs({
                 "camera": {
                     "isEnabled": True,
@@ -114,7 +117,7 @@ class DailyVision(EventHandler):
                 video_frame = self.__queue.get(timeout=5)
 
                 if video_frame:
-                    image = Image.frombytes("RGBA", (video_frame.width, video_frame.height), video_frame.buffer)
+                    image = Image.frombytes("RGB", (video_frame.width, video_frame.height), video_frame.buffer)
                     result_image = self.__pipeline.predict(params, image)
                     self.__camera.write_frame(result_image.tobytes())
                     #pil = Image.fromarray(result.render()[0], mode="RGB").tobytes()
@@ -123,9 +126,8 @@ class DailyVision(EventHandler):
 
     def on_video_frame(self, participant_id, video_frame):
         # Process ~15 frames per second (considering incoming frames at 30fps).
-        if time.time() - self.__time > 5: #0.05:
+        if time.time() - self.__time > 2: #0.05:
             self.__time = time.time()
-            self.setup_camera(video_frame)
             self.__queue.put(video_frame)    
 
     def wave(self, emoji="👋"):

app/pipeline.py

@@ -181,8 +181,8 @@ class Pipeline:
             
             self.pipe(
                 prompt="warmup",
-                image=[Image.new("RGBA", (640, 480))],
-                control_image=[Image.new("RGBA", (640, 480))],
+                image=[Image.new("RGB", (640, 480))],
+                control_image=[Image.new("RGB", (640, 480))],
         )
 
     def predict(self, params: "Pipeline.InputParams", image) -> Image.Image: