WarpGAN: Automatic Caricature Generation

We propose, WarpGAN, a fully automatic network that can generate caricatures given an input face photo. Besides transferring rich texture styles, WarpGAN learns to automatically predict a set of control points that can warp the photo into a caricature, while preserving identity. We introduce an identity-preserving adversarial loss that aids the discriminator to distinguish between different subjects. Moreover, WarpGAN allows customization of the generated caricatures by controlling the exaggeration extent and the visual styles. Experimental results on a public domain dataset, WebCaricature, show that WarpGAN is capable of generating a diverse set of caricatures while preserving the identities. Five caricature experts suggest that caricatures generated by WarpGAN are visually similar to hand-drawn ones and only prominent facial features are exaggerated.

Tattoo Image Search at Scale: Joint Detection and Compact Representation Learning

The explosive growth of digital images in video surveillance and social media has led to the significant need for efficient search of persons of interest in law enforcement and forensic applications. Despite tremendous progress in primary biometric traits (e.g., face and fingerprint) based person identification, a single biometric trait alone cannot meet the desired recognition accuracy in forensic scenarios. Tattoos, as one of the important soft biometric traits, have been found to be valuable for assisting in person identification. However, tattoo search in a large collection of unconstrained images remains a difficult problem, and existing tattoo search methods mainly focus on matching cropped tattoos, which is different from real application scenarios. To close the gap, we propose an efficient tattoo search approach that is able to learn tattoo detection and compact representation jointly in a single convolutional neural network (CNN) via multi-task learning. While the features in the backbone network are shared by both tattoo detection and compact representation learning, individual latent layers of each sub-network optimize the shared features toward the detection and feature learning tasks, respectively. We resolve the small batch size issue inside the joint tattoo detection and compact representation learning network via random image stitch and preceding feature buffering. We evaluate the proposed tattoo search system using multiple public-domain tattoo benchmarks, and a gallery set with about 300K distracter tattoo images compiled from these datasets and images from the Internet. In addition, we also introduce a tattoo sketch dataset containing 300 tattoos for sketch-based tattoo search. Experimental results show that the proposed approach has superior performance in tattoo detection and tattoo search at scale compared to several state-of-the-art tattoo retrieval algorithms.

Learning Face Age Progression: A Pyramid Architecture of GANs

The two underlying requirements of face age progression, i.e. aging accuracy and identity permanence, are not well handled in the literature. In this paper, we present a novel generative adversarial network based approach. It separately models the constraints for the intrinsic subject-specific characteristics and the age-specific facial changes with respect to the elapsed time, ensuring that the generated faces present desired aging effects while simultaneously keeping personalized properties stable. Further, to generate more lifelike facial details, high-level age-specific features conveyed by the synthesized face are estimated by a pyramidal adversarial discriminator at multiple scales, which simulates the aging effects in a finer manner. The proposed method is applicable for diverse face samples in the presence of variations in pose, expression, makeup, etc., and remarkably vivid aging effects are achieved. Both visual fidelity and quantitative evaluations show that the approach advances the state-of-the-art.

Altered Fingerprints: Detection and Localization

Fingerprint alteration, also referred to as obfuscation presentation attack, is to intentionally tamper or damage the real friction ridge patterns to avoid identification by an AFIS. This paper proposes a method for detection and localization of fingerprint alterations. Our main contributions are: (i) design and train CNN models on fingerprint images and minutiae-centered local patches in the image to detect and localize regions of fingerprint alterations, and (ii) train a Generative Adversarial Network (GAN) to synthesize altered fingerprints whose characteristics are similar to true altered fingerprints. A successfully trained GAN can alleviate the limited availability of altered fingerprint images for research. A database of 4,815 altered fingerprints from 270 subjects, and an equal number of rolled fingerprint images are used to train and test our models. The proposed approach achieves a True Detection Rate (TDR) of 99.24% at a False Detection Rate (FDR) of 2%, outperforming published results. The synthetically generated altered fingerprint dataset will be open-sourced.

DocFace+: ID Document to Selfie Matching

Numerous activities in our daily life require us to verify who we are by showing our ID documents containing face images, such as passports and driver licenses, to human operators. However, this process is slow, labor intensive and unreliable. As such, an automated system for matching ID document photos to live face images (selfies) in real time and with high accuracy is required. In this paper, we propose DocFace+ to meet this objective. We first show that gradient-based optimization methods converge slowly (due to the underfitting of classifier weights) when many classes have very few samples, a characteristic of existing ID-selfie datasets. To overcome this shortcoming, we propose a method, called dynamic weight imprinting (DWI), to update the classifier weights, which allows faster convergence and more generalizable representations. Next, a pair of sibling networks with partially shared parameters are trained to learn a unified face representation with domain-specific parameters. Cross-validation on an ID-selfie dataset shows that while a publicly available general face matcher (SphereFace) only achieves a True Accept Rate (TAR) of 59.29+-1.55% at a False Accept Rate (FAR) of 0.1% on the problem, DocFace+ improves the TAR to 97.51+-0.40%.

Automatic Latent Fingerprint Segmentation

We present a simple but effective method for automatic latent fingerprint segmentation, called SegFinNet. SegFinNet takes a latent image as an input and outputs a binary mask highlighting the friction ridge pattern. Our algorithm combines fully convolutional neural network and detection-based approaches to process the entire input latent image in one shot instead of using latent patches. Experimental results on three different latent databases (i.e. NIST SD27, WVU, and an operational forensic database) show that SegFinNet outperforms both human markup for latents and the state-of-the-art latent segmentation algorithms. We further show that this improved cropping boosts the hit rate of a latent fingerprint matcher.

Face Clustering: Representation and Pairwise Constraints

Clustering face images according to their identity has two important applications: (i) grouping a collection of face images when no external labels are associated with images, and (ii) indexing for efficient large scale face retrieval. The clustering problem is composed of two key parts: face representation and choice of similarity for grouping faces. We first propose a representation based on ResNet, which has been shown to perform very well in image classification problems. Given this representation, we design a clustering algorithm, Conditional Pairwise Clustering (ConPaC), which directly estimates the adjacency matrix only based on the similarity between face images. This allows a dynamic selection of number of clusters and retains pairwise similarity between faces. ConPaC formulates the clustering problem as a Conditional Random Field (CRF) model and uses Loopy Belief Propagation to find an approximate solution for maximizing the posterior probability of the adjacency matrix. Experimental results on two benchmark face datasets (LFW and IJB-B) show that ConPaC outperforms well known clustering algorithms such as k-means, spectral clustering and approximate rank-order. Additionally, our algorithm can naturally incorporate pairwise constraints to obtain a semi-supervised version that leads to improved clustering performance. We also propose an k-NN variant of ConPaC, which has a linear time complexity given a k-NN graph, suitable for large datasets.

DocFace: Matching ID Document Photos to Selfies

Numerous activities in our daily life, including transactions, access to services and transportation, require us to verify who we are by showing our ID documents containing face images, e.g. passports and driver licenses. An automatic system for matching ID document photos to live face images in real time with high accuracy would speedup the verification process and remove the burden on human operators. In this paper, by employing the transfer learning technique, we propose a new method, DocFace, to train a domain-specific network for ID document photo matching without a large dataset. Compared with the baseline of applying existing methods for general face recognition to this problem, our method achieves considerable improvement. A cross validation on an ID-Selfie dataset shows that DocFace improves the TAR from 61.14% to 92.77% at FAR=0.1%. Experimental results also indicate that given more training data, a viable system for automatic ID document photo matching can be developed and deployed.

On the Reconstruction of Face Images from Deep Face Templates

State-of-the-art face recognition systems are based on deep (convolutional) neural networks. Therefore, it is imperative to determine to what extent face templates derived from deep networks can be inverted to obtain the original face image. In this paper, we study the vulnerabilities of a state-of-the-art face recognition system based on template reconstruction attack. We propose a neighborly de-convolutional neural network (\textit{NbNet}) to reconstruct face images from their deep templates. In our experiments, we assumed that no knowledge about the target subject and the deep network are available. To train the \textit{NbNet} reconstruction models, we augmented two benchmark face datasets (VGG-Face and Multi-PIE) with a large collection of images synthesized using a face generator. The proposed reconstruction was evaluated using type-I (comparing the reconstructed images against the original face images used to generate the deep template) and type-II (comparing the reconstructed images against a different face image of the same subject) attacks. Given the images reconstructed from \textit{NbNets}, we show that for verification, we achieve TAR of 95.20\% (58.05\%) on LFW under type-I (type-II) attacks @ FAR of 0.1\%. Besides, 96.58\% (92.84\%) of the images reconstruction from templates of partition \textit{fa} (\textit{fb}) can be identified from partition \textit{fa} in color FERET. Our study demonstrates the need to secure deep templates in face recognition systems.

Latent Fingerprint Recognition: Role of Texture Template

We propose a texture template approach, consisting of a set of virtual minutiae, to improve the overall latent fingerprint recognition accuracy. To compensate for the lack of sufficient number of minutiae in poor quality latent prints, we generate a set of virtual minutiae. However, due to a large number of these regularly placed virtual minutiae, texture based template matching has a large computational requirement compared to matching true minutiae templates. To improve both the accuracy and efficiency of the texture template matching, we investigate: i) both original and enhanced fingerprint patches for training convolutional neural networks (ConvNets) to improve the distinctiveness of descriptors associated with each virtual minutiae, ii) smaller patches around virtual minutiae and a fast ConvNet architecture to speed up descriptor extraction, iii) reduce the descriptor length, iv) a modified hierarchical graph matching strategy to improve the matching speed, and v) extraction of multiple texture templates to boost the performance. Experiments on NIST SD27 latent database show that the above strategies can improve the matching speed from 11 ms (24 threads) per comparison (between a latent and a reference print) to only 7.7 ms (single thread) per comparison while improving the rank-1 accuracy by 8.9% against 10K gallery.