YOLOv8n Object Detection Model

๐ŸŽ“ GradicAI

AI-powered exam proctoring with real-time object detection โ€” detects phones, books, and multiple persons during online exams

3.2M
Parameters
~6 MB
Model Size
80
COCO Classes
3
Used in Proctoring
640px
Input Size

๐Ÿ” What It Detects

๐Ÿ“ฑ
Mobile Phone
COCO class #67
๐Ÿ“–
Book / Notes
COCO class #73
๐Ÿ‘ฅ
Multiple Persons
COCO class #0 (count > 1)
๐Ÿšซ
No Person
COCO class #0 (count = 0)

โšก Violation Flow

The proctoring system issues up to 3 warnings before terminating the exam.

Detection โ†’ โš ๏ธ Warning 1 โ†’ โš ๏ธ Warning 2 โ†’ ๐Ÿšซ Warning 3 โ€” Terminated

๐Ÿš€ Quick Start

bash
pip install ultralytics
python
from ultralytics import YOLO
from huggingface_hub import hf_hub_download

# Download model from Hugging Face
model_path = hf_hub_download(
    repo_id="asna14/yolov8n-proctoring",
    filename="yolov8n.pt"
)

# Load and run inference
model = YOLO(model_path)
results = model("exam_frame.jpg", conf=0.25)

# Process detections
for box in results[0].boxes:
    cls = int(box.cls[0])
    conf = float(box.conf[0])
    label = results[0].names[cls]
    print(f"Detected: {label} ({conf:.2f})")

๐Ÿ›  Part of GradicAI Platform

This model powers the proctoring feature of GradicAI โ€” a full-stack AI exam & grading platform.

๐Ÿค–
AI Grading
GPT-4o powered
๐Ÿ“น
Live Proctoring
YOLOv8 + WebSocket
๐Ÿ“
Structured Exams
Forms-style UI
๐Ÿ“š
AI Quizzes
Auto-generated