# A Call to Reflect on Evaluation Practices for Age Estimation: Comparative Analysis of the State-of-the-Art and a Unified Benchmark

Source: [A Call to Reflect on Evaluation Practices for Age Estimation: Comparative Analysis of the State-of-the-Art and a Unified Benchmark](<https://www.sota2.com/research/paper/68fa72a21e209131cea57b24>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T03:53:10.280Z

Authors: Jakub Paplham, Vojtech Franc

Published: 2023-07-10T14:02:31.000Z

## Abstract

Comparing different age estimation methods poses a challenge due to the unreliability of published results stemming from inconsistencies in the benchmarking process. Previous studies have reported continuous performance improvements over the past decade using specialized methods; however, our findings challenge these claims. This paper identifies two trivial, yet persistent issues with the currently used evaluation protocol and describes how to resolve them. We offer an extensive comparative analysis for state-of-the-art facial age estimation methods. Surprisingly, we find that the performance differences between the methods are negligible compared to the effect of other factors, such as facial alignment, facial coverage, image resolution, model architecture, or the amount of data used for pretraining. We use the gained insights to propose using FaRL as the backbone model and demonstrate its effectiveness on all public datasets. We make the source code and exact data splits public on GitHub.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Age Estimation on UTKFace Unified benchmark CVPR’24 protocol](<https://www.sota2.com/research/sota/age-estimation-on-utkface-unified-benchmark-cvpr-24-protocol.md>): Age Estimation; UTKFace Unified benchmark CVPR’24 protocol; MAE: 3.87.

- [Age estimation on CACD (test)](<https://www.sota2.com/research/sota/age-estimation-on-cacd-test.md>): Age Estimation; CACD (test); MAE: 3.96.

- [Facial Age Estimation on UTKFace](<https://www.sota2.com/research/sota/facial-age-estimation-on-utkface.md>): Facial Age Estimation; UTKFace; Mean Absolute Error (MAE): 3.87.

- [Facial Age Estimation on CACD](<https://www.sota2.com/research/sota/facial-age-estimation-on-cacd.md>): Facial Age Estimation; CACD; MAE: 3.96.

- [Facial Age Estimation on MORPH](<https://www.sota2.com/research/sota/facial-age-estimation-on-morph.md>): Facial Age Estimation; Morph; MAE: 3.04.

- [Age Estimation on FG-NET](<https://www.sota2.com/research/sota/age-estimation-on-fg-net.md>): Age Estimation; FG-NET; MAE: 4.95.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2307.04570>)

[PDF](<https://arxiv.org/pdf/2307.04570.pdf>)

[Implementation](<https://www.sota2.com/github.com/anonymous>)