RECORD
Surveillance Face Anti-spoofing
Computer Vision and Pattern Recognition · Submitted 3 Jan 2023
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim Finally, a large number of experiments verify the quality of the SuHiFiMask dataset and the superiority of the proposed CQIL.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Face Anti-spoofing (FAS) is essential to secure face recognition systems from various physical attacks. However, recent research generally focuses on short-distance applications (i.e., phone unlocking) while lacking consideration of long-distance scenes (i.e., surveillance security checks). In order to promote relevant research and fill this gap in the community, we collect a large-scale Surveillance High-Fidelity Mask (SuHiFiMask) dataset captured under 40 surveillance scenes, which has 101 subjects from different age groups with 232 3D attacks (high-fidelity masks), 200 2D attacks (posters, portraits, and screens), and 2 adversarial attacks. In this scene, low image resolution and noise interference are new challenges faced in surveillance FAS. Together with the SuHiFiMask dataset, we propose a Contrastive Quality-Invariance Learning (CQIL) network to alleviate the performance degradation caused by image quality from three aspects: (1) An Image Quality Variable module (IQV) is introduced to recover image information associated with discrimination by combining the super-resolution network. (2) Using generated sample pairs to simulate quality variance distributions to help contrastive learning strategies obtain robust feature representation under quality variation. (3) A Separate Quality Network (SQN) is designed to learn discriminative features independent of image quality. Finally, a large number of experiments verify the quality of the SuHiFiMask dataset and the superiority of the proposed CQIL.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Face Anti-spoofing (FAS) is essential to secure face recognition systems from various physical attacks. | background 1.00 | 0.04 | 0.03 | asserted 1.00 |
| s1 | However, recent research generally focuses on short-distance applications (i.e., phone unlocking) while lacking consideration of long-distance scenes (i.e., surveillance security checks). | gap 1.00 | 0.08 | 0.03 | asserted 0.99 |
| s2 | In order to promote relevant research and fill this gap in the community, we collect a large-scale Surveillance High-Fidelity Mask (SuHiFiMask) dataset captured under 40 surveillance scenes, which has 101 subjects from different age groups with 232 3D attacks (high-fidelity masks), 200 2D attacks (posters, portraits, and screens), and 2 adversarial attacks. | contribution 0.89 | 0.97 | 0.05 | asserted 1.00 |
| s3 | In this scene, low image resolution and noise interference are new challenges faced in surveillance FAS. | background 0.87 | 0.21 | 0.03 | asserted 1.00 |
| s4 | Together with the SuHiFiMask dataset, we propose a Contrastive Quality-Invariance Learning (CQIL) network to alleviate the performance degradation caused by image quality from three aspects: (1) An Image Quality Variable module (IQV) is introduced to recover image information associated with discrimination by combining the super-resolution network. | method 0.71 | 0.91 | 0.09 | asserted 0.97 |
| s5 | (2) Using generated sample pairs to simulate quality variance distributions to help contrastive learning strategies obtain robust feature representation under quality variation. | method 0.98 | 0.88 | 0.04 | asserted 0.88 |
| s6 | (3) A Separate Quality Network (SQN) is designed to learn discriminative features independent of image quality. | method 0.96 | 0.93 | 0.05 | asserted 0.99 |
| s7 | Finally, a large number of experiments verify the quality of the SuHiFiMask dataset and the superiority of the proposed CQIL. | result 0.96 | 0.94 | 0.77 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Face Anti-spoofing (FAS) is essential to secure face recognition systems from various physical attacks. However, recent research generally focuses on short-distance applications (i.e., phone unlocking) while lacking consideration of long-distance scenes (i.e., surveillance security checks). In order to promote relevant research and fill this gap in the community, we collect a large-scale Surveillance High-Fidelity Mask (SuHiFiMask) dataset captured under 40 surveillance scenes, which has 101 subjects from different age groups with 232 3D attacks (high-fidelity masks), 200 2D attacks (posters, portraits, and screens), and 2 adversarial attacks. In this scene, low image resolution and noise interference are new challenges faced in surveillance FAS. Together with the SuHiFiMask dataset, we propose a Contrastive Quality-Invariance Learning (CQIL) network to alleviate the performance degradation caused by image quality from three aspects: (1) An Image Quality Variable module (IQV) is introduced to recover image information associated with discrimination by combining the super-resolution network. (2) Using generated sample pairs to simulate quality variance distributions to help contrastive learning strategies obtain robust feature representation under quality variation. (3) A Separate Quality Network (SQN) is designed to learn discriminative features independent of image quality. Finally, a large number of experiments verify the quality of the SuHiFiMask dataset and the superiority of the proposed CQIL.
Human review
How the literature uses it
2 arXiv papers cite this in 2 sentences
Papers citing it · 1 sentence without a confident label
Cites it as background 1 paper
-
It has become an increasingly critical concern, including competition organization [ref] and algorithm design [ref] , due to its wide applications in financial payment, phone unlocking, and access control.
FM-ViT: Flexible Modal Vision Transformers for Face Anti-Spoofing · Introduction · 0.87
What it cites · 9 sentences without a confident label
Uses the method or tool of 1 paper
-
In particular, in order to effectively prevent criminals from hiding their identity information during security checks in surveillance scenes, we crafted adversarial mask [ref] and adversarial hat [ref] that can induce face recognition systems to categorize the registered identity as unknown identity, aiming to increase the challenge to algorithm stability.
Adversarial Mask: Real-World Universal Adversarial Attack on Face Recognition Model · Acquisition Details of SuHiFiMask. · 0.85
Cites as background 1 paper
-
In the case of FMeta [ref] , MT-FAS [ref] , D$^2$AM [ref] , and SDA [ref] , they aim to find the generalized feature directions via meta-learning strategies.
Generalizable Representation Learning for Mixture Domain Face Anti-Spoofing · Related Work · 0.90
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.