Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- colab-notebook
|
| 4 |
+
- vision-language-model
|
| 5 |
+
- fastvlm
|
| 6 |
+
- Apple FastVLM-0.5B
|
| 7 |
+
- video-analysis
|
| 8 |
+
- image-captioning
|
| 9 |
+
license: mit
|
| 10 |
+
language:
|
| 11 |
+
- en
|
| 12 |
+
base_model:
|
| 13 |
+
- apple/FastVLM-0.5B
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
# FastVLM-0.5B Video Analysis and Captioning
|
| 17 |
+
|
| 18 |
+
This Colab notebook demonstrates how to use the Apple FastVLM-0.5B model from Hugging Face (`apple/FastVLM-0.5B`) to perform video analysis and generate captions for video frames.
|
| 19 |
+
|
| 20 |
+
The notebook covers the following steps:
|
| 21 |
+
|
| 22 |
+
1. **Model Loading**: Loading the FastVLM-0.5B model and its processor using the Hugging Face `transformers` library.
|
| 23 |
+
2. **Image Captioning**: Testing the model on sample images.
|
| 24 |
+
3. **Video Processing**: Reading a video file (specifically `/content/drive/MyDrive/VLMs/vlm_warehouse.mp4` in this case) and extracting frames.
|
| 25 |
+
4. **Inference on Video Frames**: Running the FastVLM model on selected video frames to generate descriptions.
|
| 26 |
+
5. **Caption Overlay and Video Generation**: Creating a new video file where the original video frames are displayed with the generated captions overlaid or stacked below. The captions update based on the inference performed on key frames.
|
| 27 |
+
|
| 28 |
+
## Usage
|
| 29 |
+
|
| 30 |
+
You can open this notebook directly in Google Colab by clicking the "Open in Colab" badge on the repository page.
|
| 31 |
+
|
| 32 |
+
To run the video analysis section, make sure you have a video file available in your Google Drive at the path specified in the notebook (currently set to `/content/drive/MyDrive/VLMs/vlm_warehouse.mp4`).
|
| 33 |
+
|
| 34 |
+
## Model Details
|
| 35 |
+
|
| 36 |
+
- **Model ID**: `apple/FastVLM-0.5B`
|
| 37 |
+
- **Model Type**: Vision-Language Model
|
| 38 |
+
- **Library**: Hugging Face `transformers`
|
| 39 |
+
|
| 40 |
+
## Datasets Used
|
| 41 |
+
|
| 42 |
+
- Conceptual Captions (used for initial model testing)
|
| 43 |
+
- Custom video file (`vlm_warehouse.mp4` from Google Drive)
|
| 44 |
+
|
| 45 |
+
## Example Output
|
| 46 |
+
|
| 47 |
+
Stacked video with original frames that are available with generated captions at the bottom
|
| 48 |
+
|
| 49 |
+
## Acknowledgements
|
| 50 |
+
|
| 51 |
+
- The developers of the FastVLM-0.5B model.
|
| 52 |
+
- The Hugging Face team for the `transformers` and `huggingface_hub` libraries.
|
| 53 |
+
- Google Colab for providing the environment.
|
| 54 |
+
|
| 55 |
+
Feel free to explore and adapt this notebook for your own video analysis tasks!
|