Skip to content

kr1347/ResNet-Model

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 

Repository files navigation

Age Estimation via Fine-tuned ResNet-50

Open In Colab

Facial age estimation using a fine-tuned ResNet-50 regression model trained on the UTKFace dataset. Predicts continuous age values from face images.

Architecture

ResNet-50 (pretrained on ImageNet) with the final fully connected layer replaced by a regression head:

FC(2048 → 256) → ReLU → Dropout(0.3) → FC(256 → 1)

Dataset

UTKFace — Large-scale face dataset with age labels spanning 0–116 years. Images are labeled with age, gender, and ethnicity encoded in filenames.

  • Train/Test split: 80/20 (stratified random)
  • Input resolution: 224×224 (ResNet standard)

Training

Hyperparameter Value
Loss MSELoss
Optimizer Adam (lr=1e-4)
Epochs 30
Batch size 32
Augmentation RandomCrop, ColorJitter, HorizontalFlip

ImageNet normalization applied (mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]).

Results

Training loss monitored across 30 epochs with validation set evaluation per epoch. Transfer learning from ImageNet features significantly accelerates convergence compared to training from scratch.

Usage

Open the notebook in Colab. Mount Google Drive and provide the UTKFace dataset zip path. Run all cells sequentially.

References

  • He, K. et al. (2016). Deep Residual Learning for Image Recognition. CVPR 2016
  • Zhang, Z. et al. (2017). Age progression/regression by conditional adversarial autoencoder. CVPR 2017 (UTKFace dataset)

About

Age estimation from facial images using fine-tuned ResNet-50 — UTKFace dataset, PyTorch, MSE regression, custom augmentation pipeline

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors