Inspiration
The inspiration behind ChildGuardAI stemmed from a deep-seated desire to address the challenges faced by children in the digital age. Recognizing the prevalence of child abuse and the lack of effective reporting mechanisms, we aimed to develop a solution that leverages AI to provide immediate assistance and promote child safety.
What it does
Problems it solves:
- Comprehensive Platform: ChildGuardAI offers a holistic solution to address various challenges related to child safety in the digital realm.
- Advanced Technologies: Leveraging cutting-edge technologies, including machine learning and real-time communication, the platform delivers innovative solutions to tackle online threats.
- Detecting Explicit Content: ChildGuardAI employs machine learning models to detect and filter explicit content, ensuring a safer browsing experience for children.
- SOS Button: The platform features an SOS button for immediate assistance, providing quick access to help in emergency situations.
- Empowerment: ChildGuardAI empowers parents, educators, and caregivers with the necessary tools and resources to safeguard children online.
- Live Location Tracking: With live location tracking capabilities, the platform enables real-time monitoring of children's whereabouts, enhancing safety measures.
- Video Call Counseling: ChildGuardAI offers video call counseling services, allowing children to seek support and guidance from qualified professionals.
- Community Forum: A supportive community forum fosters collaboration and knowledge-sharing among users, promoting a collective effort towards child safety.
- Digital Literacy: Through education and awareness initiatives, ChildGuardAI promotes digital literacy among children and adults, empowering them to navigate the online world safely.
- Proactive Intervention: The platform facilitates proactive intervention through reporting mechanisms and preventive measures, mitigating risks and preventing harm to children online.
How we built it
We utilized a combination of technologies and methodologies to develop ChildGuardAI:
- Trained image classification model using ResNet50 and Fastai for detecting explicit content in images.
- Employed NLP techniques with Google BERT for text classification tasks.
- Integrated Gemini's RAG API for the chatbot functionality.
- Implemented the SOS system using Twilio for immediate assistance.
- Leveraged OpenStreetMap for live location tracking.
Challenges we ran into
- Integration of Machine Learning Models 🤖: Integrating machine learning models for detecting explicit content posed a significant challenge during ChildGuardAI's development.
- Technical Challenges with Features 💻: Setting up features like live location tracking and the SOS button presented technical challenges, requiring robust backend infrastructure and real-time communication capabilities.
- Optimization and Testing 🛠️: Overcoming these obstacles involved extensive testing and optimization efforts to ensure performance, accuracy, and user experience.
Accomplishments that we're proud of
- We are able to integrate so many features in one
What we learned
- We learned that technologies like AI can solve anything
What's next for ChildGuardAI
Our next target is to develop a mobile application and enhance real-time detection capabilities.
Built With
- bert
- deep-learning
- fastai
- gemini
- langchain
- machine-learning
- pytesseract
- python
- pytoch
- resnet18
- streamlit
- twilio
- zegocloud
Log in or sign up for Devpost to join the conversation.