Created for ShellHacks 2024 🦖🌴
1st Place Winners in @Waymo challenge
EmergencyAct uses surveillance cameras to detect accidents and emergencies in real-time, providing essential insights and support.
- Object Detection: Detects a variety of people in an area.
- Pose Estimation: Identifies if a person is in a dangerous position.
- Privacy Preservation: Maintains privacy by using tags without saving biometric data.
- Pose Coordinate Change: Utilizes LLM to determine the position of individuals.
- Accident Prevention Recommendations: Provides suggestions to enhance safety.
- Timestamp Detection: Uses Vision AI to detect the time of a fall.
- Relative Position Analysis: Determines the accident location relative to the individual.
- Cause Analysis: Analyzes potential causes based on timestamps.
- Emergency Call Prompting: Crafts messages to call emergency services using text-to-speech.
- Dashboard Summary: Displays a summary of accidents and emergencies.
We combine various tools to enhance safety in your spaces.
- 📌 Computer Vision: OpenCV
- 📌 Natural Language Processing: GPT-3.5 Turbo, prompt classification
- 📌 Generative AI: GPT-4-o, base64 image encoding
- 📌 Data Visualization: Plotly
- 📌 Model Vision YOLOv8mpose, ultralytics, pytorch
- 📌 Machine Learning Random Forest (97% test set accuracy), MLP (97% test set accuracy)
- 📌 libraries streamlit, CV2, ultralytics, numpy, base64, os, requests, openai, collections, datetime, torch, pickle, sklearn, joblib, pandas, numpy, fastapi, shutil, pymongo, urllib, aiohttp, json, PIL, gridfs, BASE64.
streamlit_app
├─ home.py
├─ .streamlit
│ └─ secrets.toml
├─ algorithms
│ └─ BB_prompt.py
│ └─ gpt_vision.py
├─ api
│ └─ main.py
├─ assets
│ └─ images
├─ features
│ └─ call_or_sms.py
│ └─ prompt.py
├─ input
│ └─ input_video.avi
│ └─ input1.avi
├─ pages
│ └─ Accident location.py
│ └─ Cause of accident.py
│ └─ Emergency call.py
│ └─ instructions.py
│ └─ Video analysis.pyDeployed with: Streamlit Cloud