OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Age Gender Estimation

Age Gender Estimation is an open-source Keras convolutional neural network implementation that detects faces in webcam or directory images and estimates age and gender. It includes a pretrained-model demo and scripts for preparing training data, training models and evaluating age estimates.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Age Gender Estimation — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you need a documented local demo or a codebase to adapt for training and evaluating age and gender estimation from face images.

Who it is for

  • Developers exploring a self-hosted face-image estimation workflow.
  • Researchers or practitioners preparing face-image datasets and training or evaluating a Keras model.

Documented uses

  • Generate age and gender estimates for faces in webcam input or a directory of images.
  • Prepare IMDB-WIKI training data by filtering noisy labels and creating a database.
  • Train a model, retain checkpoint weights and evaluate age estimates on APPA-REAL.
Before you choose

Requirements & limits

The repository is public and MIT-licensed. Its pretrained weights are subject to the IMDB-WIKI dataset’s academic-research-only conditions.

  • The demo accepts webcam input or an image directory and downloads its pretrained model automatically.
  • The documented dependency is Python 3.6 or later; the README names one tested software and hardware environment without guaranteeing compatibility elsewhere.
  • IMDB-WIKI is used for training, and its academic-research-only conditions also restrict the pretrained models.
  • The UTKFace data-preparation path is marked currently unsupported; the README says tightly cropped UTKFace images require --margin 0 when using weights trained on that dataset.
  • The README reports APPA-REAL evaluation results and distinguishes its own results from a higher-performing result that used fine-tuning on APPA-REAL training data.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Run the demo on webcam or directory images

Use the repository’s pretrained model to generate age and gender estimates for detected faces.

Before you start

  • Python 3.6 or later.
  • A webcam for webcam mode, or a directory of images for directory mode.

Steps

  1. Run the command below to start the webcam demo.

    Documented command
    python demo.py

  2. To use a directory instead, run the command below.

    Documented command
    python demo.py --image_dir [IMAGE_DIR]

Expected outcome

The demo detects faces and estimates age and gender; the pretrained model is automatically downloaded into `pretrained_models`.

Guide limitations

  • The README lists a tested environment of Ubuntu 16.04, Python 3.6.9, TensorFlow 2.3.0, CUDA 10.01 and cuDNN 7.6; this is not a compatibility guarantee.
  • Pretrained weights are restricted to academic research under the IMDB-WIKI dataset conditions.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
yu4u/age-gender-estimation · Official homepage title · retained in its source language
How we review sourcesSuggest a correction