File size: 1,502 Bytes
5369807 8277066 5369807 8277066 5369807 e4bf2b1 54f807f e4bf2b1 54f807f e4bf2b1 54f807f e4bf2b1 54f807f e4bf2b1 54f807f e4bf2b1 54f807f e4bf2b1 54f807f e4bf2b1 54f807f e4bf2b1 54f807f e4bf2b1 54f807f e4bf2b1 54f807f e4bf2b1 54f807f e4bf2b1 54f807f e4bf2b1 54f807f 8277066 | 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: 5.26.0
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: [`saja003/MuniVis`](https://huggingface.co/saja003/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 |