P-Age: Pexels Dataset for Robust Spatio-Temporal Apparent Age Classification - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2024

P-Age: Pexels Dataset for Robust Spatio-Temporal Apparent Age Classification

Abid Ali
  • Function : Author
  • PersonId : 1117945
Marisetty Ashish
  • Function : Author
Francois F Bremond


Age estimation is a challenging task that has numerous applications. In this paper, we propose a new direction for age classification that utilizes a video-based model to address challenges such as occlusions, low-resolution, and lighting conditions. To address these challenges, we propose AgeFormer which utilizes spatio-temporal information on the dynamics of the entire body dominating facebased methods for age classification. Our novel two-stream architecture uses TimeSformer and EfficientNet as backbones, to effectively capture both facial and body dynamics information for efficient and accurate age estimation in videos. Furthermore, to fill the gap in predicting age in realworld situations from videos, we construct a video dataset called Pexels Age (P-Age) for age classification. The proposed method achieves superior results compared to existing face-based age estimation methods and is evaluated in situations where the face is highly occluded, blurred, or masked. The method is also cross-tested on a variety of challenging video datasets such as Charades, Smarthome, and Thumos-14. The code and dataset is available at https://github.com/Ashish013/AgeFormer.
Fichier principal
Vignette du fichier
2377_wacv_age.pdf (4.52 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04356537 , version 1 (20-12-2023)




Abid Ali, Marisetty Ashish, Francois F Bremond. P-Age: Pexels Dataset for Robust Spatio-Temporal Apparent Age Classification. WACV 2024 - Winter Conference on Applications of Computer Vision, Jan 2024, HAWAII, United States. ⟨hal-04356537⟩
30 View
9 Download



Gmail Mastodon Facebook X LinkedIn More