arXiv subjects

Xiang Wu

Publications and source records attributed to Xiang Wu.

48 records · Page 3Linked to original sources

arXiv2018

A Light CNN for Deep Face Representation with Noisy Labels

The volume of convolutional neural network (CNN) models proposed for face recognition has been continuously growing larger to better fit large amount of training data. When training data are obtained from internet, the labels are likely to be ambiguous and inaccurate. This paper presents a Light CNN framework to learn a compact embedding on the large-scale face data with massive noisy labels. First, we introduce a variation of maxout activation, called Max-Feature-Map (MFM), into each convolutional layer of CNN. Different from maxout activation that uses many feature maps to linearly approximate an arbitrary convex activation function, MFM does so via a competitive relationship. MFM can not only separate noisy and informative signals but also play the role of feature selection between two feature maps. Second, three networks are carefully designed to obtain better performance meanwhile reducing the number of parameters and computational costs. Lastly, a semantic bootstrapping method is proposed to make the prediction of the networks more consistent with noisy labels. Experimental results show that the proposed framework can utilize large-scale noisy data to learn a Light model that is efficient in computational costs and storage spaces. The learned single network with a 256-D representation achieves state-of-the-art results on various face benchmarks without fine-tuning. The code is released on https://github.com/AlfredXiangWu/LightCNN.

cs.CV↗

arXiv2018

Learning Scene-specific Object Detectors Based on a Generative-Discriminative Model with Minimal Supervision

One object class may show large variations due to diverse illuminations, backgrounds and camera viewpoints. Traditional object detection methods often perform worse under unconstrained video environments. To address this problem, many modern approaches model deep hierarchical appearance representations for object detection. Most of these methods require a timeconsuming training process on large manual labelling sample set. In this paper, the proposed framework takes a remarkably different direction to resolve the multi-scene detection problem in a bottom-up fashion. First, a scene-specific objector is obtained from a fully autonomous learning process triggered by marking several bounding boxes around the object in the first video frame via a mouse. Here the human labeled training data or a generic detector are not needed. Second, this learning process is conveniently replicated many times in different surveillance scenes and results in particular detectors under various camera viewpoints. Thus, the proposed framework can be employed in multi-scene object detection applications with minimal supervision. Obviously, the initial scene-specific detector, initialized by several bounding boxes, exhibits poor detection performance and is difficult to improve with traditional online learning algorithm. Consequently, we propose Generative-Discriminative model to partition detection response space and assign each partition an individual descriptor that progressively achieves high classification accuracy. A novel online gradual optimized process is proposed to optimize the Generative-Discriminative model and focus on the hard samples.Experimental results on six video datasets show our approach achieves comparable performance to robust supervised methods, and outperforms the state of the art self-learning methods under varying imaging conditions.

cs.CV↗

arXiv2017

Anti-Makeup: Learning A Bi-Level Adversarial Network for Makeup-Invariant Face Verification

Makeup is widely used to improve facial attractiveness and is well accepted by the public. However, different makeup styles will result in significant facial appearance changes. It remains a challenging problem to match makeup and non-makeup face images. This paper proposes a learning from generation approach for makeup-invariant face verification by introducing a bi-level adversarial network (BLAN). To alleviate the negative effects from makeup, we first generate non-makeup images from makeup ones, and then use the synthesized non-makeup images for further verification. Two adversarial networks in BLAN are integrated in an end-to-end deep network, with the one on pixel level for reconstructing appealing facial images and the other on feature level for preserving identity information. These two networks jointly reduce the sensing gap between makeup and non-makeup images. Moreover, we make the generator well constrained by incorporating multiple perceptual losses. Experimental results on three benchmark makeup face datasets demonstrate that our method achieves state-of-the-art verification accuracy across makeup status and can produce photo-realistic non-makeup face images.

cs.CV↗

arXiv2017

Coupled Deep Learning for Heterogeneous Face Recognition

Heterogeneous face matching is a challenge issue in face recognition due to large domain difference as well as insufficient pairwise images in different modalities during training. This paper proposes a coupled deep learning (CDL) approach for the heterogeneous face matching. CDL seeks a shared feature space in which the heterogeneous face matching problem can be approximately treated as a homogeneous face matching problem. The objective function of CDL mainly includes two parts. The first part contains a trace norm and a block-diagonal prior as relevance constraints, which not only make unpaired images from multiple modalities be clustered and correlated, but also regularize the parameters to alleviate overfitting. An approximate variational formulation is introduced to deal with the difficulties of optimizing low-rank constraint directly. The second part contains a cross modal ranking among triplet domain specific images to maximize the margin for different identities and increase data for a small amount of training samples. Besides, an alternating minimization method is employed to iteratively update the parameters of CDL. Experimental results show that CDL achieves better performance on the challenging CASIA NIR-VIS 2.0 face recognition database, the IIIT-D Sketch database, the CUHK Face Sketch (CUFS), and the CUHK Face Sketch FERET (CUFSF), which significantly outperforms state-of-the-art heterogeneous face recognition methods.

cs.CV↗

arXiv2017

An silicon nanocrystal laser

Silicon lasers have been the most challenging element for the monolithic integrated Si photonics. Here we report the first successful all-Si laser at room temperature based on silicon nanocrystals (Si NCs) with high optical gains. The active Si NC layer was made from hydrogen silsesquioxane (HSQ) that had undergone a phase separation annealing, followed by high-pressure hydrogen passivation. Using the nanoimprint technique, a second-order Bragg grating was made on the active layer as the distributed feedback (DFB) laser cavity. This Si NC DFB laser device was optically pumped with femtosecond pulses. Criteria for lasing action, i.e., the threshold effect, the polarization dependence and the significant spectral narrowing of stimulated emission were evidently fulfilled, which indicated the realization of an all-Si laser.

physics.optics↗

arXiv2017

Adversarial Discriminative Heterogeneous Face Recognition

The gap between sensing patterns of different face modalities remains a challenging problem in heterogeneous face recognition (HFR). This paper proposes an adversarial discriminative feature learning framework to close the sensing gap via adversarial learning on both raw-pixel space and compact feature space. This framework integrates cross-spectral face hallucination and discriminative feature learning into an end-to-end adversarial network. In the pixel space, we make use of generative adversarial networks to perform cross-spectral face hallucination. An elaborate two-path model is introduced to alleviate the lack of paired images, which gives consideration to both global structures and local textures. In the feature space, an adversarial loss and a high-order variance discrepancy loss are employed to measure the global and local discrepancy between two heterogeneous distributions respectively. These two losses enhance domain-invariant feature learning and modality independent noise removing. Experimental results on three NIR-VIS databases show that our proposed approach outperforms state-of-the-art HFR methods, without requiring of complex network or large-scale training dataset.

cs.CV↗

arXiv2017

Attention-Set based Metric Learning for Video Face Recognition

Face recognition has made great progress with the development of deep learning. However, video face recognition (VFR) is still an ongoing task due to various illumination, low-resolution, pose variations and motion blur. Most existing CNN-based VFR methods only obtain a feature vector from a single image and simply aggregate the features in a video, which less consider the correlations of face images in one video. In this paper, we propose a novel Attention-Set based Metric Learning (ASML) method to measure the statistical characteristics of image sets. It is a promising and generalized extension of Maximum Mean Discrepancy with memory attention weighting. First, we define an effective distance metric on image sets, which explicitly minimizes the intra-set distance and maximizes the inter-set distance simultaneously. Second, inspired by Neural Turing Machine, a Memory Attention Weighting is proposed to adapt set-aware global contents. Then ASML is naturally integrated into CNNs, resulting in an end-to-end learning scheme. Our method achieves state-of-the-art performance for the task of video face recognition on the three widely used benchmarks including YouTubeFace, YouTube Celebrities and Celebrity-1000.

cs.CV↗

arXiv2017

Wasserstein CNN: Learning Invariant Features for NIR-VIS Face Recognition

Heterogeneous face recognition (HFR) aims to match facial images acquired from different sensing modalities with mission-critical applications in forensics, security and commercial sectors. However, HFR is a much more challenging problem than traditional face recognition because of large intra-class variations of heterogeneous face images and limited training samples of cross-modality face image pairs. This paper proposes a novel approach namely Wasserstein CNN (convolutional neural networks, or WCNN for short) to learn invariant features between near-infrared and visual face images (i.e. NIR-VIS face recognition). The low-level layers of WCNN are trained with widely available face images in visual spectrum. The high-level layer is divided into three parts, i.e., NIR layer, VIS layer and NIR-VIS shared layer. The first two layers aims to learn modality-specific features and NIR-VIS shared layer is designed to learn modality-invariant feature subspace. Wasserstein distance is introduced into NIR-VIS shared layer to measure the dissimilarity between heterogeneous feature distributions. So W-CNN learning aims to achieve the minimization of Wasserstein distance between NIR distribution and VIS distribution for invariant deep feature representation of heterogeneous face images. To avoid the over-fitting problem on small-scale heterogeneous face data, a correlation prior is introduced on the fully-connected layers of WCNN network to reduce parameter space. This prior is implemented by a low-rank constraint in an end-to-end network. The joint formulation leads to an alternating minimization for deep feature representation at training stage and an efficient computation for heterogeneous data at testing stage. Extensive experiments on three challenging NIR-VIS face recognition databases demonstrate the significant superiority of Wasserstein CNN over state-of-the-art methods.

cs.CV↗

arXiv2015

Learning Robust Deep Face Representation

With the development of convolution neural network, more and more researchers focus their attention on the advantage of CNN for face recognition task. In this paper, we propose a deep convolution network for learning a robust face representation. The deep convolution net is constructed by 4 convolution layers, 4 max pooling layers and 2 fully connected layers, which totally contains about 4M parameters. The Max-Feature-Map activation function is used instead of ReLU because the ReLU might lead to the loss of information due to the sparsity while the Max-Feature-Map can get the compact and discriminative feature vectors. The model is trained on CASIA-WebFace dataset and evaluated on LFW dataset. The result on LFW achieves 97.77% on unsupervised setting for single net.

cs.CV↗

arXiv2009

Molecular spring: from spider silk to silkworm silk

In this letter, we adopt a new approach combining theoretical modeling with silk stretching measurements to explore the mystery of the structures between silkworm and spider silks, leading to the differences in mechanical response against stretching. Hereby the typical stress-strain profiles are reproduced by implementing the newly discovered and verified "$β$-sheet splitting" mechanism, which primarily varies the secondary structure of protein macromolecules; our modeling and simulation results show good accordance with the experimental measurements. Hence, it can be concluded that the post-yielding mechanical behaviors of both kinds of silks are resulted from the splitting of crystallines while the high extensibility of spider dragline is attributed to the tiny $β$-sheets solely existed in spider silk fibrils. This research reveals for the first time the structural factors leading to the significant difference between spider and silkworm silks in mechanical response to the stretching force. Additionally, the combination of theoretical modeling with experiments opens up a completely new approach in resolving conformation of various biomacromolecules.

physics.bio-ph↗

arXiv2006

Interplay between evolutionary game and network structure: the coevolution of social net, cooperation and wealth

We study the interplay between evolutionary game and network structure and show how the dynamics of the game affect the growth pattern of the network and how the evolution of the network influence the cooperative behavior in the game. Simulation results show that the payoff-based preferential attachment mechanism leads to the emergence of a scale-free structural property, $P(k)\sim k^{-γ}$. Moreover, we investigate the average path length and the assortative mixing features. The obtained results indicate that the network has small-world and positive assortative behaviors, which are consistent with the observations of some real social networks. In parallel, we found that the evolution of the underlying network structure effectively promotes the cooperation level of the game. We also investigate the wealth distribution obtained by our model, which is consistent with the Pareto law in the real observation. In addition, the analysis of the generated scale-free network structure is provided for better understanding the evolutionary dynamics of our model.

physics.soc-ph↗

arXiv2005

The synchronizability of highly clustered scale-free networks

In this letter, we consider the effect of clustering coefficient on the synchronizability of coupled oscillators located on scale-free networks. The analytic result for the value of clustering coefficient aiming at a highly clustered scale-free network model, the Holme-Kim is obtained, and the relationship between network synchronizability and clustering coefficient is reported. The simulation results strongly suggest that the more clustered the network, the poorer its synchronizability.

cond-mat.stat-mech↗