2021
DOI: 10.48550/arxiv.2109.07950
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Learnable Multi-level Frequency Decomposition and Hierarchical Attention Mechanism for Generalized Face Presentation Attack Detection

Abstract: With the increased deployment of face recognition systems in our daily lives, face presentation attack detection (PAD) is attracting a lot of attention and playing a key role in securing face recognition systems. Despite the great performance achieved by the hand-crafted and deep learning based methods in intra-dataset evaluations, the performance drops when dealing with unseen scenarios. In this work, we propose a dual-stream convolution neural networks (CNNs) framework. One stream adapts four learnable frequ… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
3
0

Year Published

2022
2022
2023
2023

Publication Types

Select...
2
1

Relationship

0
3

Authors

Journals

citations
Cited by 3 publications
(3 citation statements)
references
References 26 publications
0
3
0
Order By: Relevance
“…Authors of [22] proposed a hybrid method that utilized Discriminant Correlation Analysis (DCA), Canonical Correlation Analysis (CCA) and intensity distribution control using image contrast adjustment along with transfer learning and HOG features. Fang et al [23] used a dual stream fusion model combining frequency, texture and semantic features. A multi-level frequency decomposition was also applied to address generalisation in this fusion method.…”
Section: Related Workmentioning
confidence: 99%
“…Authors of [22] proposed a hybrid method that utilized Discriminant Correlation Analysis (DCA), Canonical Correlation Analysis (CCA) and intensity distribution control using image contrast adjustment along with transfer learning and HOG features. Fang et al [23] used a dual stream fusion model combining frequency, texture and semantic features. A multi-level frequency decomposition was also applied to address generalisation in this fusion method.…”
Section: Related Workmentioning
confidence: 99%
“…A multi-level fusion strategy was followed to incorporate multiple biometric modalities. In a dual stream fusion model, Fang et al [ 22 ] used frequency domain features and complementary RGB features. This model included a hierarchical attention module as well as a multi-stage fusion strategy.…”
Section: Related Workmentioning
confidence: 99%
“…It has been seamlessly inserted in networks to incorporate frequency components to color space [6,7]. Additional frequency domain analysis explores spoofing clues in both face liveness detection [8,9] and forgery detection [10,11] tasks.…”
Section: Introductionmentioning
confidence: 99%