MiVOLO: Multi-input Transformer for Age and Gender Estimation

TL;DR

MiVOLO is a multi-input transformer model that improves age and gender estimation accuracy, especially in challenging conditions, by integrating facial and person image data and achieving state-of-the-art results.

Contribution

The paper introduces MiVOLO, a novel multi-input transformer model that combines face and person data for robust age and gender recognition, including a new benchmark dataset.

Findings

Achieves state-of-the-art performance on four benchmarks.

Outperforms human accuracy in age recognition.

Operates in real-time with high accuracy.

Abstract

Age and gender recognition in the wild is a highly challenging task: apart from the variability of conditions, pose complexities, and varying image quality, there are cases where the face is partially or completely occluded. We present MiVOLO (Multi Input VOLO), a straightforward approach for age and gender estimation using the latest vision transformer. Our method integrates both tasks into a unified dual input/output model, leveraging not only facial information but also person image data. This improves the generalization ability of our model and enables it to deliver satisfactory results even when the face is not visible in the image. To evaluate our proposed model, we conduct experiments on four popular benchmarks and achieve state-of-the-art performance, while demonstrating real-time processing capabilities. Additionally, we introduce a novel benchmark based on images from the Open Images Dataset. The ground truth annotations for this benchmark have been meticulously generated by human annotators, resulting in high accuracy answers due to the smart aggregation of votes. Furthermore, we compare our model's age recognition performance with human-level accuracy and demonstrate that it significantly outperforms humans across a majority of age ranges. Finally, we grant public access to our models, along with the code for validation and inference. In addition, we provide extra annotations for used datasets and introduce our new benchmark.