---
license: mit
tags:
- computer-vision
- object-detection
- image-classification
- multi-task-learning
- pytorch
- efficientnet
datasets:
- pascal-voc-2012
---

# CV Multitask Pipeline

Multi-task computer vision model performing simultaneous object classification and bounding box detection on PASCAL VOC 2012 (20 classes).

## Model Details

- **Architecture:** Frozen EfficientNet-B0 backbone + custom classification head + custom detection head
- **Backbone:** ImageNet-pretrained EfficientNet-B0 (frozen, 4M params)
- **Classification head:** 667K trainable params, outputs 20-class logits
- **Detection head:** 329K trainable params, outputs normalized bbox coordinates

## Training

- **Dataset:** PASCAL VOC 2012 (5,717 train / 5,823 val images)
- **Epochs:** 30 (best checkpoint at epoch 5)
- **Loss:** Weighted multi-task loss (CrossEntropy + SmoothL1, λ_cls=1.0, λ_det=5.0)
- **Hardware:** Google Colab T4 GPU

## Results

| Metric | Value |
|--------|-------|
| Top-1 Accuracy | 79.9% |
| Mean IoU | 0.468 |
| IoU@0.5 Accuracy | 49.2% |

## Usage

```python
import torch
from huggingface_hub import hf_hub_download

checkpoint_path = hf_hub_download(
    repo_id="OmUniyal/cv-multitask-pipeline",
    filename="best_model.pt"
)
checkpoint = torch.load(checkpoint_path, map_location="cpu", weights_only=True)
```

Full model code and inference pipeline: [GitHub repo](https://github.com/OmUniyal/cv-multitask-pipeline)

## Limitations

- Closed-set classification — no "none of the above" option for out-of-distribution images
- Overfitting observed after epoch 5
- Single bbox regression — doesn't handle multiple objects per image

## Live Demo

[HuggingFace Spaces](https://huggingface.co/spaces/OmUniyal/cv-multitask-pipeline)