Indian Food Classifier – TEAM ANALIZ
This model classifies images of Indian food into 95 different categories. Developed by TEAM ANALIZ, it uses a fine-tuned Vision Transformer (ViT) model to recognize a variety of Indian dishes.
Model Details
Model Description
This model is trained on a large dataset of Indian food images, designed to recognize popular Indian dishes with high accuracy. The model uses the Vision Transformer (ViT) architecture, which has proven effective for image classification tasks. The model is optimized for use in mobile and web-based food recognition applications.
- Developed by: TEAM ANALIZ
- Model type: Image Classification
- License: MIT License
- Finetuned from model: ViT (Vision Transformer)
Model Sources
- Repository: TEAM ANALIZ Repository
Uses
Direct Use
This model can be used to classify Indian food images into predefined categories. It is intended for food recognition applications, nutrition tracking apps, or any system that needs food image classification.
Out-of-Scope Use
This model is not suited for classifying images from non-food categories. It is not designed for identifying food in non-Indian cuisines or for use in situations where the model might be subjected to adversarial or misleading images.
Bias, Risks, and Limitations
The model is trained on a dataset that may not cover all possible variations of Indian food. Additionally, some food items may have multiple regional variations that are not fully represented. Users should be cautious about the model's accuracy in regions where a different variety of the dish is served.
Recommendations
Users should be aware of the limitations of this model when deploying it in real-world applications, especially when dealing with regional variations or low-quality images. Further fine-tuning may be required to improve classification for certain food items.
- Downloads last month
- -