A-LINK: Recognizing Disguised Faces via Active Learning based Inter-Domain Knowledge
Citations
5 citations
Cites background from "A-LINK: Recognizing Disguised Faces..."
...In the worst case, these changes are made intentionally to hide ones identity or imitate the appearance of another person [15, 38, 47]....
[...]
2 citations
Cites background or methods from "A-LINK: Recognizing Disguised Faces..."
...Compared to A-LINK [28], an average absolute increase of 2....
[...]
...Models trained with A2-LINK even outperform A-LINK [28] by a significant margin, thus reinforcing...
[...]
...(A-LINK [28]) give a reasonably good increase in performance, both when dealing with disguise and multiresolution as covariates....
[...]
...2A shorter version of the manuscript was presented at IEEE International Conference on BTAS, 2019 [28]....
[...]
...This paper builds on top of A-LINK2 [28] and introduces an adversarial noise component while constructing hybrid noise inputs for the algorithm....
[...]
References
108 citations
91 citations
"A-LINK: Recognizing Disguised Faces..." refers methods in this paper
...[10] uses such an approach to work with unlabeled data in the target domain....
[...]
86 citations
"A-LINK: Recognizing Disguised Faces..." refers methods in this paper
...These two techniques have been combined in the form of active-supervised domain adaptation [18] for improved performance....
[...]
76 citations
"A-LINK: Recognizing Disguised Faces..." refers background in this paper
...According to the competition, MiRA-Face [11], AEFRL [24] and UMDNets [1] are the current state-of-the-art for this dataset....
[...]
...According to the competition, MiRA-Face [11], AEFRL [24] and UMDNets [1] are the current state-of-the-art for this dataset....
[...]
...The Disguised Faces in the Wild (DFW) dataset [11, 23] dataset contains more than 11,155 images of 1000 subjects with different kinds of disguise variations....
[...]
73 citations
"A-LINK: Recognizing Disguised Faces..." refers methods in this paper
...Receiver operating characteristic (ROC) curves pertaining to the overall cases for DFW are shown in Figure 3, using DenseNet and L-CSSE....
[...]
...This L-CSSE Autoencoder is used with a Siamese network built on top of it with three fully-connected layers: the absolute difference in feature vectors is passed as input to the fully-connected layers....
[...]
...other feature extraction model, experiments are also shown using a Local Class Sparsity Supervised Autoencoder (LCSSE) [15] (Section 5....
[...]
...LCSSE autoencoder is trained on LFW before being used as a featurization model for A-LINK....
[...]
...As a proof-of-concept of the proposed approach and the boost in performance it yields, we have also conducted experiments using an L-CSSE model [15] for feature extraction for DFW....
[...]