arXiv subjects
Publications and source records attributed to Xiaotian Li.
Let $π(x;γ_1,γ_2)$ denote the number of primes $p$ with $p\leqslant x$ and $p=\lfloor n^{1/γ_1}_1\rfloor=\lfloor n^{1/γ_2}_2\rfloor$, where $\lfloor t\rfloor$ denotes the integer part of $t\in\mathbb{R}$ and $1/2<γ_2<γ_1<1$ are fixed constants. In this paper, we show that $π(x;γ_1,γ_2)$ holds an asymptotic formula for $21/11<γ_1+γ_2<2$, which constitutes an improvement upon the previous result of Baker [1].
Contrastive learning has shown promising potential for learning robust representations by utilizing unlabeled data. However, constructing effective positive-negative pairs for contrastive learning on facial behavior datasets remains challenging. This is because such pairs inevitably encode the subject-ID information, and the randomly constructed pairs may push similar facial images away due to the limited number of subjects in facial behavior datasets. To address this issue, we propose to utilize activity descriptions, coarse-grained information provided in some datasets, which can provide high-level semantic information about the image sequences but is often neglected in previous studies. More specifically, we introduce a two-stage Contrastive Learning with Text-Embeded framework for Facial behavior understanding (CLEF). The first stage is a weakly-supervised contrastive learning method that learns representations from positive-negative pairs constructed using coarse-grained activity information. The second stage aims to train the recognition of facial expressions or facial action units by maximizing the similarity between image and the corresponding text label names. The proposed CLEF achieves state-of-the-art performance on three in-the-lab datasets for AU recognition and three in-the-wild datasets for facial expression recognition.
Image restoration aims to restore high-quality images from degraded counterparts and has seen significant advancements through deep learning techniques. The technique has been widely applied to mobile devices for tasks such as mobile photography. Given the resource limitations on mobile devices, such as memory constraints and runtime requirements, the efficiency of models during deployment becomes paramount. Nevertheless, most previous works have primarily concentrated on analyzing the efficiency of single modules and improving them individually. This paper examines the efficiency across different layers. We propose a roadmap that can be applied to further accelerate image restoration models prior to deployment while simultaneously increasing PSNR (Peak Signal-to-Noise Ratio) and SSIM (Structural Similarity Index). The roadmap first increases the model capacity by adding more parameters to partial convolutions on FLOPs non-sensitive layers. Then, it applies partial depthwise convolution coupled with decoupling upsampling/downsampling layers to accelerate the model speed. Extensive experiments demonstrate that our approach decreases runtime by up to 13% and reduces the number of parameters by up to 23%, while increasing PSNR and SSIM on several image restoration datasets. Source Code of our method is available at \href{https://github.com/xiangyu8/MOFA}{https://github.com/xiangyu8/MOFA}.
Recent studies have focused on utilizing multi-modal data to develop robust models for facial Action Unit (AU) detection. However, the heterogeneity of multi-modal data poses challenges in learning effective representations. One such challenge is extracting relevant features from multiple modalities using a single feature extractor. Moreover, previous studies have not fully explored the potential of multi-modal fusion strategies. In contrast to the extensive work on late fusion, there are limited investigations on early fusion for channel information exploration. This paper presents a novel multi-modal reconstruction network, named Multimodal Channel-Mixing (MCM), as a pre-trained model to learn robust representation for facilitating multi-modal fusion. The approach follows an early fusion setup, integrating a Channel-Mixing module, where two out of five channels are randomly dropped. The dropped channels then are reconstructed from the remaining channels using masked autoencoder. This module not only reduces channel redundancy, but also facilitates multi-modal learning and reconstruction capabilities, resulting in robust feature learning. The encoder is fine-tuned on a downstream task of automatic facial action unit detection. Pre-training experiments were conducted on BP4D+, followed by fine-tuning on BP4D and DISFA to assess the effectiveness and robustness of the proposed framework. The results demonstrate that our method meets and surpasses the performance of state-of-the-art baseline methods.
Visual localization is critical to many applications in computer vision and robotics. To address single-image RGB localization, state-of-the-art feature-based methods match local descriptors between a query image and a pre-built 3D model. Recently, deep neural networks have been exploited to regress the mapping between raw pixels and 3D coordinates in the scene, and thus the matching is implicitly performed by the forward pass through the network. However, in a large and ambiguous environment, learning such a regression task directly can be difficult for a single network. In this work, we present a new hierarchical scene coordinate network to predict pixel scene coordinates in a coarse-to-fine manner from a single RGB image. The proposed method, which is an extension of HSCNet, allows us to train compact models which scale robustly to large environments. It sets a new state-of-the-art for single-image localization on the 7-Scenes, 12 Scenes, Cambridge Landmarks datasets, and the combined indoor scenes.
In laser-based free-space optical communication, besides OAM beams, Hermite-Gaussian (HG) modes or HG-mode coherently composed states (HG-MCCS) can also be adopted as the information carrier to extend the channel capacity with the spatial pattern based encoding and decoding link. The light field of HG-MCCS is mainly determined by three independent parameters, including indexes of HG modes, relative initial phases between two eigenmodes, and scale coefficients of the eigenmodes, which can obtain a large number of effective coding modes at a low mode order. The beam intensity distributions of the HG-MCCSs have obvious distinguishable spatial characteristics and can keep propagation invariance, which are convenient to be decoded by the convolutional neural network (CNN) based image recognition method. We experimentally utilize HG-MCCS to realize a communication link including encoding, transmission under atmospheric turbulence (AT), and decoding based on CNN. With the index order of eigenmodes within six, 125 HG-MCCS are generated and used for information encoding, and the average recognition accuracy reached 99.5% for non-AT conditions. For the 125-level color images transmission, the error rate of the system is less than 1.8% even under the weak AT condition. Our work provides a useful basis for the future combination of dense data communication and artificial intelligence technology.
For several emerging technologies such as augmented reality, autonomous driving and robotics, visual localization is a critical component. Directly regressing camera pose/3D scene coordinates from the input image using deep neural networks has shown great potential. However, such methods assume a stationary data distribution with all scenes simultaneously available during training. In this paper, we approach the problem of visual localization in a continual learning setup -- whereby the model is trained on scenes in an incremental manner. Our results show that similar to the classification domain, non-stationary data induces catastrophic forgetting in deep networks for visual localization. To address this issue, a strong baseline based on storing and replaying images from a fixed buffer is proposed. Furthermore, we propose a new sampling method based on coverage score (Buff-CS) that adapts the existing sampling strategies in the buffering process to the problem of visual localization. Results demonstrate consistent improvements over standard buffering methods on two challenging datasets -- 7Scenes, 12Scenes, and also 19Scenes by combining the former scenes.
Recent studies on the automatic detection of facial action unit (AU) have extensively relied on large-sized annotations. However, manually AU labeling is difficult, time-consuming, and costly. Most existing semi-supervised works ignore the informative cues from the temporal domain, and are highly dependent on densely annotated videos, making the learning process less efficient. To alleviate these problems, we propose a deep semi-supervised framework Knowledge-Spreader (KS), which differs from conventional methods in two aspects. First, rather than only encoding human knowledge as constraints, KS also learns the Spatial-Temporal AU correlation knowledge in order to strengthen its out-of-distribution generalization ability. Second, we approach KS by applying consistency regularization and pseudo-labeling in multiple student networks alternately and dynamically. It spreads the spatial knowledge from labeled frames to unlabeled data, and completes the temporal information of partially labeled video clips. Thus, the design allows KS to learn AU dynamics from video clips with only one label allocated, which significantly reduce the requirements of using annotations. Extensive experiments demonstrate that the proposed KS achieves competitive performance as compared to the state of the arts under the circumstances of using only 2% labels on BP4D and 5% labels on DISFA. In addition, we test it on our newly developed large-scale comprehensive emotion database, which contains considerable samples across well-synchronized and aligned sensor modalities for easing the scarcity issue of annotations and identities in human affective computing. The new database will be released to the research community.
Emotion is an experience associated with a particular pattern of physiological activity along with different physiological, behavioral and cognitive changes. One behavioral change is facial expression, which has been studied extensively over the past few decades. Facial behavior varies with a person's emotion according to differences in terms of culture, personality, age, context, and environment. In recent years, physiological activities have been used to study emotional responses. A typical signal is the electroencephalogram (EEG), which measures brain activity. Most of existing EEG-based emotion analysis has overlooked the role of facial expression changes. There exits little research on the relationship between facial behavior and brain signals due to the lack of dataset measuring both EEG and facial action signals simultaneously. To address this problem, we propose to develop a new database by collecting facial expressions, action units, and EEGs simultaneously. We recorded the EEGs and face videos of both posed facial actions and spontaneous expressions from 29 participants with different ages, genders, ethnic backgrounds. Differing from existing approaches, we designed a protocol to capture the EEG signals by evoking participants' individual action units explicitly. We also investigated the relation between the EEG signals and facial action units. As a baseline, the database has been evaluated through the experiments on both posed and spontaneous emotion recognition with images alone, EEG alone, and EEG fused with images, respectively. The database will be released to the research community to advance the state of the art for automatic emotion recognition.
Visual attention has been extensively studied for learning fine-grained features in both facial expression recognition (FER) and Action Unit (AU) detection. A broad range of previous research has explored how to use attention modules to localize detailed facial parts (e,g. facial action units), learn discriminative features, and learn inter-class correlation. However, few related works pay attention to the robustness of the attention module itself. Through experiments, we found neural attention maps initialized with different feature maps yield diverse representations when learning to attend the identical Region of Interest (ROI). In other words, similar to general feature learning, the representational quality of attention maps also greatly affects the performance of a model, which means unconstrained attention learning has lots of randomnesses. This uncertainty lets conventional attention learning fall into sub-optimal. In this paper, we propose a compact model to enhance the representational and focusing power of neural attention maps and learn the "inter-attention" correlation for refined attention maps, which we term the "Self-Diversified Multi-Channel Attention Network (SMA-Net)". The proposed method is evaluated on two benchmark databases (BP4D and DISFA) for AU detection and four databases (CK+, MMI, BU-3DFE, and BP4D+) for facial expression recognition. It achieves superior performance compared to the state-of-the-art methods.
Selecting influential nonlinear interactive features from ultrahigh dimensional data has been an important task in various fields. However, statistical accuracy and computational feasibility are the two biggest concerns when more than half a million features are collected in practice. Many extant feature screening approaches are either focused on only main effects or heavily rely on heredity structure, hence rendering them ineffective in a scenario presenting strong interactive but weak main effects. In this article, we propose a new interaction screening procedure based on joint cumulant (named JCI-SIS). We show that the proposed procedure has strong sure screening consistency and is theoretically sound to support its performance. Simulation studies designed for both continuous and categorical predictors are performed to demonstrate the versatility and practicability of our JCI-SIS method. We further illustrate the power of JCI-SIS by applying it to screen 27,554,602,881 interaction pairs involving 234,754 single nucleotide polymorphisms (SNPs) for each of the 4,000 subjects collected from polycystic ovary syndrome (PCOS) patients and healthy controls.
Fully-supervised CNN-based approaches for learning local image descriptors have shown remarkable results in a wide range of geometric tasks. However, most of them require per-pixel ground-truth keypoint correspondence data which is difficult to acquire at scale. To address this challenge, recent weakly- and self-supervised methods can learn feature descriptors from relative camera poses or using only synthetic rigid transformations such as homographies. In this work, we focus on understanding the limitations of existing self-supervised approaches and propose a set of improvements that combined lead to powerful feature descriptors. We show that increasing the search space from in-pair to in-batch for hard negative mining brings consistent improvement. To enhance the discriminativeness of feature descriptors, we propose a coarse-to-fine method for mining local hard negatives from a wider search space by using global visual image descriptors. We demonstrate that a combination of synthetic homography transformation, color augmentation, and photorealistic image stylization produces useful representations that are viewpoint and illumination invariant. The feature descriptors learned by the proposed approach perform competitively and surpass their fully- and weakly-supervised counterparts on various geometric benchmarks such as image-based localization, sparse feature matching, and image retrieval.
Vehicle trajectory optimization is essential to ensure vehicles travel efficiently and safely. This paper presents an infrastructure assisted constrained connected automated vehicles (CAVs) trajectory optimization method on curved roads. This paper systematically formulates the problem based on a curvilinear coordinate which is flexible to model complex road geometries. Further, to deal with the spatial varying road obstacles, traffic regulations, and geometric characteristics, two-dimensional vehicle kinematics is given in a spatial formulation with exact road information provided by the infrastructure. Consequently, we applied a multi-objective model predictive control (MPC) approach to optimize the trajectories in a rolling horizon while satisfying the collision avoidances and vehicle kinematics constraints. To verify the efficiency of our method, a numerical simulation is conducted. As the results suggest, the proposed method can provide smooth vehicular trajectories, avoid road obstacles, and simultaneously follow traffic regulations, which is robust to road geometries and disturbances.
Camera pose estimation in large-scale environments is still an open question and, despite recent promising results, it may still fail in some situations. The research so far has focused on improving subcomponents of estimation pipelines, to achieve more accurate poses. However, there is no guarantee for the result to be correct, even though the correctness of pose estimation is critically important in several visual localization applications,such as in autonomous navigation. In this paper we bring to attention a novel research question, pose confidence estimation,where we aim at quantifying how reliable the visually estimated pose is. We develop a novel confidence measure to fulfil this task and show that it can be flexibly applied to different datasets,indoor or outdoor, and for various visual localization pipelines.We also show that the proposed techniques can be used to accomplish a secondary goal: improving the accuracy of existing pose estimation pipelines. Finally, the proposed approach is computationally light-weight and adds only a negligible increase to the computational effort of pose estimation.
Visual localization is critical to many applications in computer vision and robotics. To address single-image RGB localization, state-of-the-art feature-based methods match local descriptors between a query image and a pre-built 3D model. Recently, deep neural networks have been exploited to regress the mapping between raw pixels and 3D coordinates in the scene, and thus the matching is implicitly performed by the forward pass through the network. However, in a large and ambiguous environment, learning such a regression task directly can be difficult for a single network. In this work, we present a new hierarchical scene coordinate network to predict pixel scene coordinates in a coarse-to-fine manner from a single RGB image. The network consists of a series of output layers, each of them conditioned on the previous ones. The final output layer predicts the 3D coordinates and the others produce progressively finer discrete location labels. The proposed method outperforms the baseline regression-only network and allows us to train compact models which scale robustly to large environments. It sets a new state-of-the-art for single-image RGB localization performance on the 7-Scenes, 12-Scenes, Cambridge Landmarks datasets, and three combined scenes. Moreover, for large-scale outdoor localization on the Aachen Day-Night dataset, we present a hybrid approach which outperforms existing scene coordinate regression methods, and reduces significantly the performance gap w.r.t. explicit feature matching methods.
Fresh water scarcity is one of the critical challenges for global sustainable development. Several novel water resources such as passive seawater solar desalination and atmospheric water harvesting have made some progress in recent years. However, no investigation has referred to harvesting water from shallow subsurface soils, which are potential huge water reservoirs. Here, we introduce a method of solar-driven water harvesting from soils, which can provide cheap fresh water in impoverished, arid and decentralized areas. The concentrated solar energy is used to heat the soils to evaporate the soil moisture. Then vapors flow to the condenser through tubes and condense as freshwater. Sustainable water harvesting is realized by water migration due to capillary pumping effect within soils. In the laboratory condition, an experimental setup is designed and its water-harvesting ability from soils is investigated. The maximum water mass harvesting rate was 99.8 g h-1. In about 12 h, the total harvesting water could be as high as about 900 ml. The whole process is solar-driven and spontaneous without other mechanical or electrical ancillaries. The water harvesting rate under one sun energy flux (1 kW m-2) is estimated to be about 360 g h-1 with a 1 m2 solar concentrator. Our proposal provides a potential onsite and sustainable fresh water supply solution to deal with the water scarcity problem.
Image-based camera relocalization is an important problem in computer vision and robotics. Recent works utilize convolutional neural networks (CNNs) to regress for pixels in a query image their corresponding 3D world coordinates in the scene. The final pose is then solved via a RANSAC-based optimization scheme using the predicted coordinates. Usually, the CNN is trained with ground truth scene coordinates, but it has also been shown that the network can discover 3D scene geometry automatically by minimizing single-view reprojection loss. However, due to the deficiencies of the reprojection loss, the network needs to be carefully initialized. In this paper, we present a new angle-based reprojection loss, which resolves the issues of the original reprojection loss. With this new loss function, the network can be trained without careful initialization, and the system achieves more accurate results. The new loss also enables us to utilize available multi-view constraints, which further improve performance.
Image-based localization, or camera relocalization, is a fundamental problem in computer vision and robotics, and it refers to estimating camera pose from an image. Recent state-of-the-art approaches use learning based methods, such as Random Forests (RFs) and Convolutional Neural Networks (CNNs), to regress for each pixel in the image its corresponding position in the scene's world coordinate frame, and solve the final pose via a RANSAC-based optimization scheme using the predicted correspondences. In this paper, instead of in a patch-based manner, we propose to perform the scene coordinate regression in a full-frame manner to make the computation efficient at test time and, more importantly, to add more global context to the regression process to improve the robustness. To do so, we adopt a fully convolutional encoder-decoder neural network architecture which accepts a whole image as input and produces scene coordinate predictions for all pixels in the image. However, using more global context is prone to overfitting. To alleviate this issue, we propose to use data augmentation to generate more data for training. In addition to the data augmentation in 2D image space, we also augment the data in 3D space. We evaluate our approach on the publicly available 7-Scenes dataset, and experiments show that it has better scene coordinate predictions and achieves state-of-the-art results in localization with improved robustness on the hardest frames (e.g., frames with repeated structures).