Duplicate from convergingmachine/vcl-face-understanding

Co-authored-by: Mehmet Can Yavuz <convergingmachine@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,101 @@
+```yaml
+---
+language: en
+license: mit
+library_name: pytorch
+tags:
+- face-recognition
+- self-supervised-learning
+- contrastive-learning
+datasets:
+- YFCC-CelebA
+- CelebA
+---
+```
+# VCL: Variational Contrastive Learning for Face Understanding
+
+VCL is a robust self-supervised learning method designed specifically for face understanding tasks, combining variational contrastive learning with beta-divergence to effectively handle noisy and unlabeled datasets[1].
+
+## Model Details
+
+### Model Description
+
+**Developed by:** Mehmet Can Yavuz and Berrin Yanikoglu
+**Model type:** Self-Supervised Variational Contrastive Learning with Applications to Face Understanding
+**Language(s):** Python
+**License:** MIT
+**Model:** ResNet10t
+
+## Uses
+
+### Direct Use
+
+The model is designed for:
+- Face attribute recognition
+- Face verification tasks
+- Multi-label classification problems
+- Learning from noisy and unlabeled datasets
+
+## Model Architecture
+
+The architecture consists of three main components:
+- Feature extraction backbone (ResNet10t or VGG11bn)
+- Gaussian sampling head for distribution learning
+- Contrastive learning framework with augmentations
+
+## Training Details
+
+### Training Data
+
+The model was pretrained on the YFCC-CelebA dataset and you can fine-tune on CelebA dataset.
+
+### Training Procedure 
+
+#### Training Hyperparameters
+
+**Training regime:**
+- Optimizer: AdamW
+- Learning rate: 1e-3
+- Weight decay: 0.01
+- Batch size: 128
+- Temperature: 0.07
+- Beta: 0.005
+
+## Evaluation
+
+### Results
+
+Performance on CelebA test set with different pretraining approaches:
+
+| Setting | ResNet10t (1%) | VGG11bn (1%) | ResNet10t (10%) | VGG11bn (10%) |
+|---------|----------------|---------------|-----------------|---------------|
+| VCL | 0.5836 | 0.5719 | 0.6848 | 0.6796 |
+| VCL (beta) | 0.5998 | 0.5958 | 0.7098 | 0.6998 |
+
+## How to Get Started with the Model
+
+```python
+# Installation
+git clone https://github.com/convergingmachine/VCL
+cd VCL
+pip install -r requirements.txt
+
+# Training
+python train_beta.py
+
+
+## Citation
+
+```bibtex
+@INPROCEEDINGS{10582001,
+  author={Yavuz, Mehmet Can and Yanikoglu, Berrin},
+  booktitle={2024 IEEE 18th International Conference on Automatic Face and Gesture Recognition (FG)}, 
+  title={Self-Supervised Variational Contrastive Learning with Applications to Face Understanding}, 
+  year={2024},
+  pages={1-9},
+  doi={10.1109/FG59268.2024.10582001}}
+```
+
+## Model Card Contact
+
+For questions about this model, please open an issue in the GitHub repository.

yfcc_resnet10_pretrained.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c5ab72f9baf7b64e3821d020df9d9e348b9f7e47bf7e5a720a054176304c9ff
+size 67090607