Face anti-spoofing with multifeature videolet aggregation
Citations
349 citations
Cites methods from "Face anti-spoofing with multifeatur..."
...The remaining methods in Table 3 are some state-of-the-art methods which utilize handcrafted features along with a classifier [39, 9, 7]....
[...]
264 citations
183 citations
Cites background from "Face anti-spoofing with multifeatur..."
...To overcome some of these difficulties, researchers tackle the problem in different domains, such as HSV and YCbCr color space [24,25], temporal domain [26,27,28,29] and Fourier spectrum [30]....
[...]
151 citations
145 citations
Cites background or methods from "Face anti-spoofing with multifeatur..."
...for the SMAD and performance is compared with two existing algorithms [25], [54]....
[...]
...For comparison, eight recent and stateof-the-art algorithms ( [13], [54], [55], [56], [57], [58], [25]) are selected and the results are reported directly from the published papers, except in few cases as marked in Table VI....
[...]
...Many algorithms utilize features which may be heavily pertinent to the kind of attack being detected, for instance, motion [15]–[21], texture [22]–[25], reflectance properties [26], [27], or image quality [28]....
[...]
...We have implemented an existing algorithm [25] for performance comparison and Table VII shows the HTER values of the proposed and existing algorithms....
[...]
References
680 citations
"Face anti-spoofing with multifeatur..." refers background or methods in this paper
...• The CASIA-FASD dataset (CASIA) [21] consists of 600 videos corresponding to 50 subjects, separated as 240 videos in training and 360 videos in testing....
[...]
...Similarly, at the Inter Feature Fusion stage, the correlation of 0.51, 0.62, and 0.66 is observed for CASIA, MSU, and 3DMAD datasets, respectively....
[...]
...For CASIA, the improvement is relatively smaller (from 16.80 to 9.81%)....
[...]
...MSU dataset contains a higher fraction of replay attack videos compared to CASIA....
[...]
...The face anti-spoofing problem is extensively studied in literature, particularly with the introduction of Print Attack dataset [1], Replay Attack dataset [5], CASIA-FASD spoofing dataset [21], 3DMAD database [7], and MSU mobile face spoofing database [20]....
[...]
628 citations
Additional excerpts
...As opposed to Määttä et al.[13] that computes overlapping local histograms of LBPu28,1, resulting in a feature vector of size 833; multi-LBP computes global histograms at three scales, thereby resulting in a descriptor of size 361 (i.e. 59+59+243)....
[...]
...As opposed to Määttä et al.[13] that computes overlapping local histograms of LBP 8,1, resulting in a feature vector of size 833; multi-LBP computes global histograms at three scales, thereby resulting in a descriptor of size 361 (i....
[...]
611 citations
"Face anti-spoofing with multifeatur..." refers background in this paper
...In contrast, temporal evidence techniques encode spatial and temporal evidence across videos for cues such as signs of vitality [16], for spoofing detection....
[...]
610 citations
353 citations