Modelpublic

andreysnosov89-ship-it/drawing-detection-yolov8

sourceHugging Faceagpl-3.0updated 3mo agoView on Hugging Face

0likes6downloads

Model Card

YOLOv8n Drawing Detection Model

This model is a fine-tuned version of the ultralytics YOLOv8 Nano (yolov8n) model for detecting elements and objects within drawings/sketches.

Model Details

—Model type: Object Detection (YOLOv8 architecture)
—Base Model: ultralytics/yolov8n (Nano version for real-time performance)
—Library: Ultralytics
—License: AGPL-3.0

Intended Use

This model is intended for computer vision applications where identifying bounding boxes and specific classes of objects within hand-drawn sketches or digital drawings is required.

Training Parameters

The model was trained using the Ultralytics framework with the following hyper-parameters:

—Epochs: 100
—Image Size: 640x640
—Batch Size: 8
—Workers: 4
—Optimizer: Auto-selected by YOLOv8
—Early Stopping Patience: 50 epochs

Hardware Environment

—PyTorch Version: 2.7.1+cu118
—Hardware: NVIDIA RTX A4000 (15.6 GB VRAM)

How to use

You can load this model and run inference using the ultralytics package:

python

from ultralytics import YOLO

# Load the model
model = YOLO("hf_hub_username/drawing-detection-yolov8")

# Perform object detection on an image
results = model("path/to/your/drawing.jpg")

# Show results
results[0].show()