Updated

app.py

@@ -1,9 +1,6 @@
 import cv2
-import mediapipe as mp
 import numpy as np
 import google.generativeai as genai
-from langchain_core.messages import HumanMessage
-from langchain_google_genai import ChatGoogleGenerativeAI
 import os
 import streamlit as st
 from PIL import Image, ImageDraw
@@ -11,14 +8,7 @@ from PIL import Image, ImageDraw
 # Set up environment variables and configurations
 genai.configure(api_key=os.environ['GOOGLE_API_KEY'])
 
-# Set up MediaPipe (not used in the current implementation but kept for reference)
-mp_hands = mp.solutions.hands
-mp_drawing = mp.solutions.drawing_utils
-hands = mp_hands.Hands()
-
 # Initialize session state for drawing
-if 'drawing' not in st.session_state:
-    st.session_state.drawing = False
 if 'points' not in st.session_state:
     st.session_state.points = []
 
@@ -36,60 +26,59 @@ def main():
 
     with col1:
         # Create a canvas for drawing
+        canvas_image = Image.new("RGB", (800, 600), (0, 0, 0))
+
+        # Display the canvas
         canvas_placeholder = st.empty()
-        result_placeholder = st.empty()
-        if 'canvas_image' not in st.session_state:
-            st.session_state.canvas_image = Image.new("RGB", (800, 600), (0, 0, 0))
+
+        # Instructions and buttons
+        st.write("**Draw your equation below:**")
         if st.button("Reset"):
             st.session_state.points = []
-            st.session_state.canvas_image = Image.new("RGB", (800, 600), (0, 0, 0))
         if st.button("Submit"):
-            drawing_canvas = st.session_state.canvas_image.copy()
-            draw_on_image(drawing_canvas, st.session_state.points)
-            drawing_canvas.save('drawing.png')
+            draw_on_image(canvas_image, st.session_state.points)
+            canvas_image.save('drawing.png')
 
             # Send the image to Gemini and get the response
             response = send_to_gemini('drawing.png')
-            result_placeholder.text_area("Result:", value=response, height=300)
+            st.session_state.result = response
 
-        canvas_image = np.array(st.session_state.canvas_image)
-        cv2.putText(canvas_image, "Draw here", (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 1, (255, 255, 255), 2)
-        canvas_placeholder.image(canvas_image)
+        # Interactive canvas for drawing
+        st.write("**Use your mouse to draw:**")
+        mouse_down = st.button("Hold to Draw")
+        if mouse_down:
+            st.session_state.points.append(st.mouse_position())
 
-        # Handle mouse events for drawing
-        def update_canvas(image):
-            if st.session_state.drawing:
-                current_pos = st.session_state.current_pos
-                if st.session_state.new_stroke:
-                    st.session_state.points.append([])
-                    st.session_state.new_stroke = False
-                st.session_state.points[-1].append(current_pos)
-                draw_on_image(image, st.session_state.points)
+        # Draw on the image
+        draw_on_image(canvas_image, st.session_state.points)
 
-        st.session_state.drawing = st.button("Start Drawing", key="start")
-        st.session_state.current_pos = st.mouse("Mouse Position")
-        st.session_state.new_stroke = st.button("New Stroke", key="new_stroke")
-        update_canvas(st.session_state.canvas_image)
+        # Display the updated image
+        canvas_placeholder.image(np.array(canvas_image))
 
     with col2:
         st.header("Instructions")
         st.write("1. Use the left side to draw your equation.")
         st.write("2. Click **Submit** to process the drawing.")
         st.write("3. The result will be displayed below after submission.")
+        if 'result' in st.session_state:
+            st.write("**Result:**")
+            st.text_area("", value=st.session_state.result, height=300)
 
-def send_to_gemini(drawing_canvas):
+def send_to_gemini(drawing_path):
     llm = ChatGoogleGenerativeAI(model="gemini-1.5-flash-latest")
+    with open(drawing_path, 'rb') as img_file:
+        image_content = img_file.read()
     message = HumanMessage(
         content=[
             {
                 "type": "text",
                 "text": "Give me the answer of any mathematical representation in the image with the complete solution, and does not say the image contains etc.",
             },
-            {"type": "image_url", "image_url": drawing_canvas},
+            {"type": "image", "image": image_content},
         ]
     )
     response = llm.invoke([message]).content
     return response
 
 if __name__ == "__main__":
-    main()
+    main()