fernhill/fernhill-seg-b

Overview

Semantic segmentation for street and indoor scenes with 40 classes.

fernhill/fernhill-seg-b is a 90M-parameter image segmentation model, published in safetensors under the apache-2.0 license.

Intended use

• Image segmentation in the domains described above.
• Research, prototypes and products that keep a person in the loop.
• Fine-tuning as a starting point for a narrower task.

How to use

# Command-line client (planned; the shape may change)
pace pull fernhill/fernhill-seg-b

# Python (planned)
from openpace import load
model = load("fernhill/fernhill-seg-b")

Limitations

• Accuracy falls on lighting, angles and object types it rarely saw.
• Do not use it for decisions about people, such as identification or surveillance.

License

Released under apache-2.0. Read the license file before using the weights commercially.