Update app.py

app.py

@@ -92,44 +92,69 @@ def analyze_construction_media(media):
                 image = Image.open(file.name)
                 resized_image = resize_image(image)
                 image_data_url = f"data:image/png;base64,{encode_image(resized_image)}"
-                content_type = "image"
+                messages = [
+                    {
+                        "role": "user",
+                        "content": [
+                            {
+                                "type": "text",
+                                "text": f"{instruction}\n\nAnalyze this image (File {i+1}/{len(media)}). First, determine if it's a construction site. If it is, explain the image in detail, focusing on safety aspects. If it's not, briefly describe what you see."
+                            },
+                            {
+                                "type": "image_url",
+                                "image_url": {
+                                    "url": image_data_url
+                                }
+                            }
+                        ]
+                    }
+                ]
+                completion = client.chat.completions.create(
+                    model="llama-3.2-90b-vision-preview",
+                    messages=messages,
+                    temperature=0.7,
+                    max_tokens=1000,
+                    top_p=1,
+                    stream=False,
+                    stop=None
+                )
+                result = completion.choices[0].message.content
+                results.append((f"Image {i+1} analysis", result))
             elif file_type in ['mp4', 'avi', 'mov', 'wmv']:
                 # Handle video
                 frames = extract_frames_from_video(file.name)
-                image_data_url = f"data:image/png;base64,{encode_image(frames[0])}"  # Use first frame
-                content_type = "video"
-            else:
-                results.append((f"File {i+1} analysis", f"Unsupported file type: {file_type}"))
-                continue
-
-            messages = [
-                {
-                    "role": "user",
-                    "content": [
+                for j, frame in enumerate(frames):
+                    image_data_url = f"data:image/png;base64,{encode_image(frame)}"
+                    messages = [
                         {
-                            "type": "text",
-                            "text": f"{instruction}\n\nAnalyze this {content_type} (File {i+1}/{len(media)}). First, determine if it's a construction site. If it is, explain the {content_type} in detail, focusing on safety aspects. If it's not, briefly describe what you see."
-                        },
-                        {
-                            "type": "image_url",
-                            "image_url": {
-                                "url": image_data_url
-                            }
+                            "role": "user",
+                            "content": [
+                                {
+                                    "type": "text",
+                                    "text": f"{instruction}\n\nAnalyze this frame from a video (File {i+1}/{len(media)}, Frame {j+1}/{len(frames)}). First, determine if it's a construction site. If it is, explain what you observe, focusing on safety aspects. If it's not, briefly describe what you see."
+                                },
+                                {
+                                    "type": "image_url",
+                                    "image_url": {
+                                        "url": image_data_url
+                                    }
+                                }
+                            ]
                         }
                     ]
-                }
-            ]
-            completion = client.chat.completions.create(
-                model="llama-3.2-90b-vision-preview",
-                messages=messages,
-                temperature=0.7,
-                max_tokens=1000,
-                top_p=1,
-                stream=False,
-                stop=None
-            )
-            result = completion.choices[0].message.content
-            results.append((f"File {i+1} analysis", result))
+                    completion = client.chat.completions.create(
+                        model="llama-3.2-90b-vision-preview",
+                        messages=messages,
+                        temperature=0.7,
+                        max_tokens=1000,
+                        top_p=1,
+                        stream=False,
+                        stop=None
+                    )
+                    result = completion.choices[0].message.content
+                    results.append((f"Video {i+1}, Frame {j+1} analysis", result))
+            else:
+                results.append((f"File {i+1} analysis", f"Unsupported file type: {file_type}"))
 
         logger.info("Analysis completed successfully")
         return results