Protecting Young Minds Through Intelligent Content Analysis
Hackathon Demo Video: link
graph TD
A[Mac Client] -->|Screenshot every 5s| B[FastAPI Server]
B --> C[Llama 3.2 90B Vision]
C -->|Scene Description| D[Llama 3.1 8B]
D -->|Risk Assessment| E[(Firebase DB)]
E --> F[React Dashboard]
style A fill:#d4e6f1 color:#000000
style B fill:#d5f5e3 color:#000000
style C fill:#fadbd8 color:#000000
style D fill:#fadbd8 color:#000000
style E fill:#fdebd0 color:#000000
style F fill:#d6eaf8 color:#000000
graph LR
subgraph Short["Short-term Impact"]
A1[Risk Detection] --> A2[Prevention] --> A3[Communication]
end
subgraph Mid["Mid-term Impact"]
B1[School Safety] --> B2[Mental Health] --> B3[Digital Literacy]
end
subgraph Long["Long-term Impact"]
C1[Research Impact] --> C2[Policy Change] --> C3[Global Reach]
end
Short --> Mid --> Long
classDef default color:#000000
classDef boxStyle fill:#666666,stroke:#333,stroke-width:2px,color:#ffffff
classDef short fill:#d4e6f1,color:#000000
classDef mid fill:#d5f5e3,color:#000000
classDef long fill:#fadbd8,color:#000000
class Short,Mid,Long boxStyle
class A1,A2,A3 short
class B1,B2,B3 mid
class C1,C2,C3 long
graph TD
subgraph Tech["Technical Metrics"]
A1[Precision] & A2[Recall] & A3[F1 Score] & A4[Latency] --> TS[Technical Score]
end
subgraph User["User Metrics"]
B1[Parent Feedback] & B2[Adoption Rate] --> US[User Score]
end
subgraph Impact["Impact Metrics"]
C1[Safety Incidents] & C2[Mental Health] --> IS[Impact Score]
end
TS & US & IS --> TOTAL[Total Effectiveness Score]
classDef default color:#000000
classDef sectionStyle fill:#666666,stroke:#333,stroke-width:2px,color:#ffffff
classDef techStyle fill:#d4e6f1,color:#000000
classDef userStyle fill:#d5f5e3,color:#000000
classDef impactStyle fill:#fadbd8,color:#000000
classDef scoreStyle fill:#fdebd0,color:#000000
classDef totalStyle fill:#d6eaf8,color:#000000
class Tech,User,Impact sectionStyle
class A1,A2,A3,A4 techStyle
class B1,B2,B3 userStyle
class C1,C2,C3 impactStyle
class TS,US,IS scoreStyle
class TOTAL totalStyle
flowchart TD
subgraph CL["Client Layer"]
MAC[MacOS Client] --- WIN[Windows Client]
WIN --- MOB[Mobile Client]
end
subgraph ML["Model Layer"]
direction TB
API[Threat Assessment API]
VM[Vision Model]
TM[Text Model]
BM[Behavioral Model]
API --> VM & TM & BM
VM & TM & BM --> RA[Risk Aggregator]
end
subgraph DL["Data Layer"]
AUTH[Authentication] --- DB[(Database)]
DB --- LOGS[Audit Logs]
end
subgraph UL["User Interface"]
DASH[Dashboard] --- ALERT[Alert System]
ALERT --- REP[Reports]
end
CL --> API
RA --> DL
DL --> UL
%% Styles
classDef section fill:#2c3e50,color:#ffffff,stroke:#2c3e50
classDef client fill:#3498db,color:#ffffff,stroke:#2980b9
classDef model fill:#2ecc71,color:#ffffff,stroke:#27ae60
classDef data fill:#e74c3c,color:#ffffff,stroke:#c0392b
classDef ui fill:#9b59b6,color:#ffffff,stroke:#8e44ad
class CL,ML,DL,UL section
class MAC,WIN,MOB client
class API,VM,TM,BM,RA model
class AUTH,DB,LOGS data
class DASH,ALERT,REP ui
linkStyle default stroke:#ffffff,stroke-width:2px