# FeatherNets: Convolutional Neural Networks as Light as Feather for Face Anti-spoofing

arXiv: https://arxiv.org/abs/1904.09290 · HTML: https://mlsanity.com/paper/1904.09290/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Apr 2019.

## Built on by (most cited first, 14 of 14)

- [Deep Learning for Face Anti-Spoofing: A Survey](https://mlsanity.com/paper/2106.14948/) (2021)
- [Rethinking Vision Transformer and Masked Autoencoder in Multimodal Face Anti-Spoofing](https://mlsanity.com/paper/2302.05744/) (2023)
- [Digital and Physical Face Attacks: Reviewing and One Step Further](https://mlsanity.com/paper/2209.14692/) (2022)
- [Facial Depth and Normal Estimation using Single Dual-Pixel Camera](https://mlsanity.com/paper/2111.12928/) (2021)
- [Aurora Guard: Reliable Face Anti-Spoofing via Mobile Lighting System](https://mlsanity.com/paper/2102.00713/) (2021)
- [Visual Prompt Flexible-Modal Face Anti-Spoofing](https://mlsanity.com/paper/2307.13958/) (2023)
- [M3FAS: An Accurate and Robust MultiModal Mobile Face Anti-Spoofing System](https://mlsanity.com/paper/2301.12831/) (2023)
- [Review of Face Presentation Attack Detection Competitions](https://mlsanity.com/paper/2112.11290/) (2021)
- [Learning Polysemantic Spoof Trace: A Multi-Modal Disentanglement Network for Face Anti-spoofing](https://mlsanity.com/paper/2212.03943/) (2022)
- [Hyp-OC: Hyperbolic One Class Classification for Face Anti-Spoofing](https://mlsanity.com/paper/2404.14406/) (2024)
- [Learning Knowledge-based Prompts for Robust 3D Mask Presentation Attack Detection](https://mlsanity.com/paper/2505.03610/) (2025)
- [A Dataset and Benchmark Towards Multi-Modal Face Anti-Spoofing Under Surveillance Scenarios](https://mlsanity.com/paper/2103.15409/) (2021)
- [Online Adaptive Personalization for Face Anti-spoofing](https://mlsanity.com/paper/2207.12272/) (2022)
- [More than just an auxiliary loss: Anti-spoofing Backbone Training via Adversarial Pseudo-depth Generation](https://mlsanity.com/paper/2101.00200/) (2021)

## Abstract (arXiv)

Face Anti-spoofing gains increased attentions recently in both academic and industrial fields. With the emergence of various CNN based solutions, the multi-modal(RGB, depth and IR) methods based CNN showed better performance than single modal classifiers. However, there is a need for improving the performance and reducing the complexity. Therefore, an extreme light network architecture(FeatherNet A/B) is proposed with a streaming module which fixes the weakness of Global Average Pooling and uses less parameters. Our single FeatherNet trained by depth image only, provides a higher baseline with 0.00168 ACER, 0.35M parameters and 83M FLOPS. Furthermore, a novel fusion procedure with ``ensemble + cascade'' structure is presented to satisfy the performance preferred use cases. Meanwhile, the MMFD dataset is collected to provide more attacks and diversity to gain better generalization. We use the fusion method in the Face Anti-spoofing Attack Detection Challenge@CVPR2019 and got the result of 0.0013(ACER), 0.999(TPR@FPR=10e-2), 0.998(TPR@FPR=10e-3) and 0.9814(TPR@FPR=10e-4).