Change detection (CD) is to decouple object changes (i.e., object missing or appearing) from background changes (i.e., environment variations) like light and season variations in two images captured in the same scene over a long time span, presenting critical applications in disaster management, urban development, etc. In particular, the endless patterns of background changes require detectors to have a high generalization against unseen environment variations, making this task significantly challenging. Recent deep learning-based methods develop novel network architectures or optimization strategies with paired-training examples, which do not handle the generalization issue explicitly and require huge manual pixel-level annotation efforts. In this work, for the first attempt in the CD community, we study the generalization issue of CD from the perspective of data augmentation and develop a novel weakly supervised training algorithm that only needs image-level labels. Different from general augmentation techniques for classification, we propose the background-mixed augmentation that is specifically designed for change detection by augmenting examples under the guidance of a set of background-changing images and letting deep CD models see diverse environment variations. Moreover, we propose the augmented & real data consistency loss that encourages the generalization increase significantly. Our method as a general framework can enhance a wide range of existing deep learning-based detectors. We conduct extensive experiments in two public datasets and enhance four state-of-the-art methods, demonstrating the advantages of our method. We release the code at https://github.com/tsingqguo/bgmix.
It has been witnessed that masked image modeling (MIM) has shown a huge potential in self-supervised learning in the past year. Benefiting from the universal backbone vision transformer, MIM learns self-supervised visual representations through masking a part of patches of the image while attempting to recover the missing pixels. Most previous works mask patches of the image randomly, which underutilizes the semantic information that is beneficial to visual representation learning. On the other hand, due to the large size of the backbone, most previous works have to spend much time on pre-training. In this paper, we propose \textbf{Attention-driven Masking and Throwing Strategy} (AMT), which could solve both problems above. We first leverage the self-attention mechanism to obtain the semantic information of the image during the training process automatically without using any supervised methods. Masking strategy can be guided by that information to mask areas selectively, which is helpful for representation learning. Moreover, a redundant patch throwing strategy is proposed, which makes learning more efficient. As a plug-and-play module for masked image modeling, AMT improves the linear probing accuracy of MAE by $2.9\% \sim 5.9\%$ on CIFAR-10/100, STL-10, Tiny ImageNet, and ImageNet-1K, and obtains an improved performance with respect to fine-tuning accuracy of MAE and SimMIM. Moreover, this design also achieves superior performance on downstream detection and segmentation tasks. Code is available at https://github.com/guijiejie/AMT.
We apply classical statistical methods in conjunction with the state-of-the-art machine learning techniques to develop a hybrid interpretable model to analyse 454,897 online customers' behavior for a particular product category at the largest online retailer in China, that is JD. While most mere machine learning methods are plagued by the lack of interpretability in practice, our novel hybrid approach will address this practical issue by generating explainable output. This analysis involves identifying what features and characteristics have the most significant impact on customers' purchase behavior, thereby enabling us to predict future sales with a high level of accuracy, and identify the most impactful variables. Our results reveal that customers' product choice is insensitive to the promised delivery time, but this factor significantly impacts customers' order quantity. We also show that the effectiveness of various discounting methods depends on the specific product and the discount size. We identify product classes for which certain discounting approaches are more effective and provide recommendations on better use of different discounting tools. Customers' choice behavior across different product classes is mostly driven by price, and to a lesser extent, by customer demographics. The former finding asks for exercising care in deciding when and how much discount should be offered, whereas the latter identifies opportunities for personalized ads and targeted marketing. Further, to curb customers' batch ordering behavior and avoid the undesirable Bullwhip effect, JD should improve its logistics to ensure faster delivery of orders.
This paper presents a technique to train a robot to perform kick-motion in AI soccer by using reinforcement learning (RL). In RL, an agent interacts with an environment and learns to choose an action in a state at each step. When training RL algorithms, a problem called the curse of dimensionality (COD) can occur if the dimension of the state is high and the number of training data is low. The COD often causes degraded performance of RL models. In the situation of the robot kicking the ball, as the ball approaches the robot, the robot chooses the action based on the information obtained from the soccer field. In order not to suffer COD, the training data, which are experiences in the case of RL, should be collected evenly from all areas of the soccer field over (theoretically infinite) time. In this paper, we attempt to use the relative coordinate system (RCS) as the state for training kick-motion of robot agent, instead of using the absolute coordinate system (ACS). Using the RCS eliminates the necessity for the agent to know all the (state) information of entire soccer field and reduces the dimension of the state that the agent needs to know to perform kick-motion, and consequently alleviates COD. The training based on the RCS is performed with the widely used Deep Q-network (DQN) and tested in the AI Soccer environment implemented with Webots simulation software.
Current trend in neurosciences is to use naturalistic stimuli, such as cinema, class-room biology or video gaming, aiming to understand the brain functions during ecologically valid conditions. Naturalistic stimuli recruit complex and overlapping cognitive, emotional and sensory brain processes. Brain oscillations form underlying mechanisms for such processes, and further, these processes can be modified by expertise. Human cortical oscillations are often analyzed with linear methods despite brain as a biological system is highly nonlinear. This study applies a relatively robust nonlinear method, Higuchi fractal dimension (HFD), to classify cortical oscillations of math experts and novices when they solve long and complex math demonstrations in an EEG laboratory. Brain imaging data, which is collected over a long time span during naturalistic stimuli, enables the application of data-driven analyses. Therefore, we also explore the neural signature of math expertise with machine learning algorithms. There is a need for novel methodologies in analyzing naturalistic data because formulation of theories of the brain functions in the real world based on reductionist and simplified study designs is both challenging and questionable. Data-driven intelligent approaches may be helpful in developing and testing new theories on complex brain functions. Our results clarify the different neural signature, analyzed by HFD, of math experts and novices during complex math and suggest machine learning as a promising data-driven approach to understand the brain processes in expertise and mathematical cognition.
A novel speech feature fusion algorithm with independent vector analysis (IVA) and parallel convolutional neural network (PCNN) is proposed for text-independent speaker recognition. Firstly, some different feature types, such as the time domain (TD) features and the frequency domain (FD) features, can be extracted from a speaker's speech, and the TD and the FD features can be considered as the linear mixtures of independent feature components (IFCs) with an unknown mixing system. To estimate the IFCs, the TD and the FD features of the speaker's speech are concatenated to build the TD and the FD feature matrix, respectively. Then, a feature tensor of the speaker's speech is obtained by paralleling the TD and the FD feature matrix. To enhance the dependence on different feature types and remove the redundancies of the same feature type, the independent vector analysis (IVA) can be used to estimate the IFC matrices of TD and FD features with the feature tensor. The IFC matrices are utilized as the input of the PCNN to extract the deep features of the TD and FD features, respectively. The deep features can be integrated to obtain the fusion feature of the speaker's speech. Finally, the fusion feature of the speaker's speech is employed as the input of a deep convolutional neural network (DCNN) classifier for speaker recognition. The experimental results show the effectiveness and performances of the proposed speaker recognition system.
Deep Neural Networks (DNNs) have been ubiquitously adopted in internet of things and are becoming an integral of our daily life. When tackling the evolving learning tasks in real world, such as classifying different types of objects, DNNs face the challenge to continually retrain themselves according to the tasks on different edge devices. Federated continual learning is a promising technique that offers partial solutions but yet to overcome the following difficulties: the significant accuracy loss due to the limited on-device processing, the negative knowledge transfer caused by the limited communication of non-IID data, and the limited scalability on the tasks and edge devices. In this paper, we propose FedKNOW, an accurate and scalable federated continual learning framework, via a novel concept of signature task knowledge. FedKNOW is a client side solution that continuously extracts and integrates the knowledge of signature tasks which are highly influenced by the current task. Each client of FedKNOW is composed of a knowledge extractor, a gradient restorer and, most importantly, a gradient integrator. Upon training for a new task, the gradient integrator ensures the prevention of catastrophic forgetting and mitigation of negative knowledge transfer by effectively combining signature tasks identified from the past local tasks and other clients' current tasks through the global model. We implement FedKNOW in PyTorch and extensively evaluate it against state-of-the-art techniques using popular federated continual learning benchmarks. Extensive evaluation results on heterogeneous edge devices show that FedKNOW improves model accuracy by 63.24% without increasing model training time, reduces communication cost by 34.28%, and achieves more improvements under difficult scenarios such as large numbers of tasks or clients, and training different complex networks.
We present HandAvatar, a novel representation for hand animation and rendering, which can generate smoothly compositional geometry and self-occlusion-aware texture. Specifically, we first develop a MANO-HD model as a high-resolution mesh topology to fit personalized hand shapes. Sequentially, we decompose hand geometry into per-bone rigid parts, and then re-compose paired geometry encodings to derive an across-part consistent occupancy field. As for texture modeling, we propose a self-occlusion-aware shading field (SelF). In SelF, drivable anchors are paved on the MANO-HD surface to record albedo information under a wide variety of hand poses. Moreover, directed soft occupancy is designed to describe the ray-to-surface relation, which is leveraged to generate an illumination field for the disentanglement of pose-independent albedo and pose-dependent illumination. Trained from monocular video data, our HandAvatar can perform free-pose hand animation and rendering while at the same time achieving superior appearance fidelity. We also demonstrate that HandAvatar provides a route for hand appearance editing. Project website: https://seanchenxy.github.io/HandAvatarWeb.
The deaf and hard of hearing community relies on American Sign Language (ASL) as their primary mode of communication, but communication with others who do not know ASL can be difficult, especially during emergencies where no interpreter is available. As an effort to alleviate this problem, research in computer vision based real time ASL interpreting models is ongoing. However, most of these models are hand shape (gesture) based and lack the integration of facial cues, which are crucial in ASL to convey tone and distinguish similar looking signs. Thus, the integration of facial cues in computer vision based ASL interpreting models has the potential to improve performance and reliability. In this paper, we introduce a new facial expression-based classification model that can be used to improve ASL interpreting models. This model utilizes the relative angles of facial landmarks with principal component analysis and a Random Forest Classification tree model to classify frames taken from videos of ASL users signing a complete sentence. The model classifies the frames as statements or assertions. The model was able to achieve an accuracy of 82%.
Recently, probabilistic denoising diffusion models (DDMs) have greatly advanced the generative power of neural networks. DDMs, inspired by non-equilibrium thermodynamics, have not only been used for 2D image generation, but can also readily be applied to 3D point clouds. However, representing 3D shapes as point clouds has a number of drawbacks, most obvious perhaps that they have no notion of topology or connectivity. Here, we explore an alternative route and introduce tetrahedral diffusion models, an extension of DDMs to tetrahedral partitions of 3D space. The much more structured 3D representation with space-filling tetrahedra makes it possible to guide and regularize the diffusion process and to apply it to colorized assets. To manipulate the proposed representation, we develop tetrahedral convolutions, down- and up-sampling kernels. With those operators, 3D shape generation amounts to learning displacement vectors and signed distance values on the tetrahedral grid. Our experiments confirm that Tetrahedral Diffusion yields plausible, visually pleasing and diverse 3D shapes, is able to handle surface attributes like color, and can be guided at test time to manipulate the resulting shapes.