Model Card for Model sbrzz/cultural-arts-shield-v0

This modelcard provides information about sbrzz/cultural-arts-shield-v0. This model is based on MobileNetV2 and it is fine-tuned on a private dataset from www.cultural-arts.com

Model Details

This model is the really first version of ai-shield used in cultural-arts.com. It allows us to discard malecious images upload to our service before entiring in the core of the system. We fined tuned a customized MobileNetV2 in order to have only two output classes ("Accepted" and "Garbage"). We accept an image when it is related to a monument, an historical place or an historical building. In all other cases we say garbage!

Model Description

  • Developed by: cultural-arts.com AI Team
  • Model type: Customized MobileNetV2
  • License: Check the LICENCE file in the repo
  • Finetuned from model: MobileNetV2 (alpha .35)

Uses

The model, sbrzz/cultural-arts-shield-v0, is intended to serve as an initial screening mechanism for images uploaded to the cultural-arts.com service.

Direct Use

Its primary purpose is to filter out potentially malicious or irrelevant content before such images enter in a image based system. The model accept 224x224 3-channels RGB as input and categorizes images into two classes: "Accepted" and "Garbage".

Bias, Risks, and Limitations [COMING SOON]

Recommendations [COMING SOON]

How to Get Started with the Model [COMING SOON]

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.