File size: 1,510 Bytes
5ac4ae0 fc958e1 5ac4ae0 2bed17a 5ac4ae0 fc958e1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 | ---
title: Street Descriptions using MuniVis
emoji: 🛣️
colorFrom: indigo
colorTo: blue
sdk: gradio
sdk_version: 5.41.1
app_file: app.py
pinned: false
---
# 🛣️ Street Descriptions using MuniVis Model
This Space provides a simple Gradio interface for generating image captions using a fine-tuned BLIP model hosted at [`saja003/MuniVis`](https://huggingface.co/saja003/MuniVis). It also includes the option to translate the generated English captions into Arabic using a MarianMT translation model.
---
## 🔍 Features
- 🧠 Image captioning with a fine-tuned BLIP model.
- 🌐 Language selection: English or Arabic.
- 📸 Upload any street or urban image to generate a descriptive caption.
- 🤖 Translation powered by [Helsinki-NLP/opus-mt-en-ar](https://huggingface.co/Helsinki-NLP/opus-mt-en-ar).
---
## 🚀 How it works
1. Upload an image.
2. Select a language (English or Arabic).
3. Receive a descriptive caption of the scene.
- If Arabic is selected, the caption is translated automatically.
---
## 🧠 Models Used
- **BLIP (Bootstrapping Language Image Pretraining):**
- Fine-tuned on street and infrastructure images.
- Model repo: [`Projectt123/MuniVis`](https://huggingface.co/Projectt123/MuniVis)
- **MarianMT Translation Model:**
- [`Helsinki-NLP/opus-mt-en-ar`](https://huggingface.co/Helsinki-NLP/opus-mt-en-ar) for English to Arabic translation.
---
## 🛠️ Requirements
This Space uses the following libraries:
```bash
transformers
torch
gradio
Pillow |