DAF:re: A Challenging, Crowd-Sourced, Large-Scale, Long-Tailed Dataset For Anime Character Recognition
This work addresses the problem of anime character recognition for researchers and developers, providing a new dataset and framework, but it is incremental as it builds on existing methods.
The authors tackled anime character recognition by introducing DAF:re, a large-scale dataset with nearly 500,000 images across over 3,000 classes, and tested models like ResNets and ViT, finding new insights into ViT generalization and transfer learning in this domain.
In this work we tackle the challenging problem of anime character recognition. Anime, referring to animation produced within Japan and work derived or inspired from it. For this purpose we present DAF:re (DanbooruAnimeFaces:revamped), a large-scale, crowd-sourced, long-tailed dataset with almost 500 K images spread across more than 3000 classes. Additionally, we conduct experiments on DAF:re and similar datasets using a variety of classification models, including CNN based ResNets and self-attention based Vision Transformer (ViT). Our results give new insights into the generalization and transfer learning properties of ViT models on substantially different domain datasets from those used for the upstream pre-training, including the influence of batch and image size in their training. Additionally, we share our dataset, source-code, pre-trained checkpoints and results, as Animesion, the first end-to-end framework for large-scale anime character recognition: https://github.com/arkel23/animesion