Discriminative Invariant Kernel Features: A Bells-and-Whistles-Free Approach to Unsupervised Face Recognition and Pose Estimation

Pal, Dipan K.; Juefei-Xu, Felix; Savvides, Marios

doi:10.1109/cvpr.2016.603

Cited by 19 publications

(21 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Under the two relaxations, the invariance invoked to these arbitrary transformations in an NPTN would only be approximate. Nonetheless and consistent with previous work, we find in our experiments that despite the approximation, there is much to be gained overall and the invariance invoked suffices in practice as also found by (Liao, Leibo, and Poggio 2013;Pal, Juefei-Xu, and Savvides 2016).…”

Section: Non-parametric Transformation Networksupporting

confidence: 93%

“…For methods that do not enforce these conditions (unitary group conditions) in theory, no test generalization claim can be made. However, a number of studies have observed approximate albeit sufficient invariances in practice under this setting (Anselmi et al 2013;Pal, Juefei-Xu, and Savvides 2016;Pal et al 2017;Liao, Leibo, and Poggio 2013). The main motive for modelling transformations as unitary groups was to provide a theoretical connection to ConvNets and other methods that enforce other kinds of unitary invariance such as rotation invariance (Li et al 2017;Wu, Hu, and Kong 2015).…”

Section: The Transformation Network Paradigmmentioning

confidence: 99%

“…Based off the core idea of measuring moments of a group invariant distribution, multiple works had demonstrated efficacy of the ideas in more challenging real-world problems such as face recognition, though not in a neural network setting. See (Liao, Leibo, and Poggio 2013;Pal, Juefei-Xu, and Savvides 2016;Pal et al 2017).…”

Section: Prior Artmentioning

confidence: 99%

“…A number of ways exist to achieve this. One can present transformed versions of the training data as in (Niyogi, Girosi, and Poggio 1998), minimize auxiliary objectives promoting invariance during training as in (Hadsell, Chopra, and LeCun 2006) or pool over transformed versions of the representation itself as in (Liao, Leibo, and Poggio 2013;Pal, Juefei-Xu, and Savvides 2016;Pal et al 2017).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Non-Parametric Transformation Networks for Learning General Invariances from Data

Pal

Savvides

2019

AAAI

Self Cite

View full text Add to dashboard Cite

ConvNets, through their architecture, only enforce invariance to translation. In this paper, we introduce a new class of deep convolutional architectures called Non-Parametric Transformation Networks (NPTNs) which can learn general invariances and symmetries directly from data. NPTNs are a natural generalization of ConvNets and can be optimized directly using gradient descent. Unlike almost all previous works in deep architectures, they make no assumption regarding the structure of the invariances present in the data and in that aspect are flexible and powerful. We also model ConvNets and NPTNs under a unified framework called Transformation Networks (TN), which yields a better understanding of the connection between the two. We demonstrate the efficacy of NPTNs on data such as MNIST with extreme transformations and CIFAR10 where they outperform baselines, and further outperform several recent algorithms on ETH-80. They do so while having the same number of parameters. We also show that they are more effective than ConvNets in modelling symmetries and invariances from data, without the explicit knowledge of the added arbitrary nuisance transformations. Finally, we replace ConvNets with NPTNs within Capsule Networks and show that this enables Capsule Nets to perform even better.

show abstract

Section: Non-parametric Transformation Networksupporting

confidence: 93%

Section: The Transformation Network Paradigmmentioning

confidence: 99%

Section: Prior Artmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Non-Parametric Transformation Networks for Learning General Invariances from Data

Pal

Savvides

2019

AAAI

Self Cite

View full text Add to dashboard Cite

show abstract

“…In future work, we plan to propose new methods that can remove the artifacts in fake images. Moreover, other shallow methods such as the ones based on advanced correlation filters [8,10,13,17,21,40,58,60] are also potentially viable solutions to this problem, which we intend to explore further.…”

Section: Discussionmentioning

confidence: 99%

FakePolisher: Making DeepFakes More Detection-Evasive by Shallow Reconstruction

Huang

Juefei-Xu

Wang

et al. 2020

Proceedings of the 28th ACM International Conference on Multimedia

Self Cite

View full text Add to dashboard Cite

At this moment, GAN-based image generation methods are still imperfect, whose upsampling design has limitations in leaving some certain artifact patterns in the synthesized image. Such artifact patterns can be easily exploited (by recent methods) for difference detection of real and GAN-synthesized images. However, the existing detection methods put much emphasis on the artifact patterns, which can become futile if such artifact patterns were reduced. Towards reducing the artifacts in the synthesized images, in this paper, we devise a simple yet powerful approach termed FakePolisher that performs shallow reconstruction of fake images through a learned linear dictionary, intending to effectively and efficiently reduce the artifacts introduced during image synthesis. In particular, we first train a dictionary model to capture the patterns of real images. Based on this dictionary, we seek the representation of DeepFake images in a low dimensional subspace through linear projection or sparse coding. Then, we are able to perform shallow reconstruction of the 'fake-free' version of the DeepFake image, which largely reduces the artifact patterns DeepFake introduces. The comprehensive evaluation on 3 state-of-the-art DeepFake detection methods and fake images generated by 16 popular GAN-based fake image generation techniques, demonstrates the effectiveness of our technique. Overall, through reducing artifact patterns, our technique significantly reduces the accuracy of the 3 state-of-theart fake image detection methods, i.e., 47% on average and up to 93% in the worst case. Our results confirm the limitation of current fake detection methods and calls the attention of DeepFake researchers and practitioners for more general-purpose fake detection techniques. CCS CONCEPTS • Security and privacy → Human and societal aspects of security and privacy; • Computing methodologies → Computer vision.

show abstract