Latest state-of-the-art detec-tors, including single (SSD [12] and YOLO [16]) or mul-tiple stage RCNN-like [17] neural networks, are based on In order to meet the needs of real-time monitoring on mobile devices, the message have observed the amazing achievements of VGG16 in the 2014 Imagenet image classification competition and the application advantages of convolutional neural network in various fields such as image classification, image segmentation and target detection. In the detection output of SSD, the position of the detected objects is shown with the bounding boxes and, for each bounding box, the class Jul 9, 2018 · The network does not look at the complete image. This paper examines more closely how object detection has evolved in the era of deep learning over the past years. 15:750639. Object detection is a computer vision task that aims to locate objects in digital images. The process of object detection is getting faster and better with development in the fields of neural networks and deep learning domains. In this article we combine FPGA and neural networks technologies to solve the real-time object recognition problem. Dec 5, 2013 · This paper presents a simple and yet powerful formulation of object detection as a regression problem to object bounding box masks, and defines a multi-scale inference procedure which is able to produce high-resolution object detections at a low cost by a few network applications. Automotive embedded algorithms have very high constraints in terms of latency, accuracy and power consumption. The task involves identifying the position and boundaries of objects in an image, and classifying the objects into different categories. Neural Network for Satellite Data Classification Using Tensorflow in Python-> A step-by-step guide for Landsat 5 multispectral data classification for binary built-up/non-built-up class prediction, with repo Jun 22, 2020 · Hey Adrian, if I have a Convolutional Neural Network trained for image classification, how in the world am I going to use it for object detection? Based on your explanation above, it seems like image classification and object detection are fundamentally different, requiring two different types of network architectures. Instead, parts of the image which have high probabilities of containing the object. , Eda Kavlakoglu. At the same time, the demand for neural network computing 2. The intuition behind developing this model was the fact that previous frameworks needed fixed size input images, a constraint that was artificial. Faster R-CNN is the latest model in this family of object detector algorithms. The image features learned through deep learning techniques are more Detecting deforestation from satellite images-> using FastAI and ResNet50, with repo fsdl_deforestation_detection. Code. However, these deep neural networks are trained on hundreds of classes and require a Mar 11, 2021 · The architecture consists of two neural networks — Detector and Classifier. These systems involve not only recognizing and classifying every object in an image, but localizing each one by drawing the Jun 21, 2022 · a classification task — classifies an object (person vs car etc. & Li, G. Front. 750639 Oct 7, 2019 · YOLO-based Convolutional Neural Network family of models for object detection and the most recent variation called YOLOv3. Keywords: seeing, vision, object recognition, brain, deep neural network, eye tracking, saliency map. Fast R-CNN operates by using a single shared CNN to extract features from the entire image. The network is an extension of the Faster-RCNN network where we add an additional input network branch for processing the depth image. The training strategies and pipelines dramatically vary among different models. It’s in your phone, computer, car, camera, and more. This article discusses the self-driving cars’ vision systems, role of DL to interpret complex vision, enhance perception, and actuate kinematic manoeuvres in self-driving cars [8]. It forms a crucial part of vision recognition, alongside Aug 11, 2016 · DIGITS 4 introduces a new object detection workflow that allows you to train networks to detect objects (such as faces, vehicles, or pedestrians) in images and define bounding boxes around them. Mar 9, 2020 · In this paper, we propose a binarized neural network learning method called BiDet for efficient object detection. How to use a pre-trained YOLOv3 to perform object localization and detection on new photographs. Jan 1, 2022 · In [79] the authors presented a new framework for deep convolutional neural networks aimed at the task of object detection. With the rise of autonomous vehicles, smart video surveillance, facial detection and various people counting applications, fast and accurate object detection systems are rising in demand. It contains 170 images with 345 instances of pedestrians, and we will use it to illustrate how to use the Utilizing a graph neural network on the raw radar tensor we gain a significant improvement of +10% in average precision over a grid-based convolutional baseline network. Keras is a Python library for deep learning that wraps the powerful numerical libraries Theano and TensorFlow. 1 Region-Based Convolutional Neural Network (R-CNN) R-CNN, a short form of region-based convolutional neural network, is one of the most widely used object detection model that falls under two-stage object detectors. 8ms per image. Jul 9, 2020 · Object detection is an important component of computer vision. A detector is an Object Detection Neural Network. However, how to design a directly-trained SNN for the regression task of object detection still remains a challenging problem Jul 12, 2019 · Convolutional Neural Network. 638182 Nov 8, 2022 · Object detection using convolutional neural networks addresses the recognition problem solely in terms of feature extraction and disregards knowledge and experience to explore higher-level relationships between objects. The MS-CNN consists of a proposal sub-network and a detection sub-network. This network can be integrated into any object Contributors: Jacob Murel Ph. To improve the performance Sep 16, 2022 · Three-dimensional object detection is crucial for autonomous driving to understand the driving environment. See the post Deep Learning for Object Detection with DIGITS for a walk-through of how to use this new functionality. face recognition, not face classification). Conventional network binarization methods directly quantize the weights and activations in one-stage or two-stage detectors with constrained representational capacity, so that the information redundancy in the networks causes numerous false positives and degrades the performance Oct 23, 2020 · 3. Object detection algorithms can be based on traditional Computer Vision approaches or Neural Network ones [1]. In the object detection and recognition field, neural networks are in use for a decade but became prominent due to the improvement of hardware new techniques for training these networks on large datasets . In this paper, a comprehensive study is carried out on deep learning and its major component named convolution neural networks (CNN) that are largely used for object detection. For this tutorial, we will be finetuning a pre-trained Mask R-CNN model on the Penn-Fudan Database for Pedestrian Detection and Segmentation. Deep learning is a powerful machine learning technique in which the object detector automatically learns image features required for detection tasks. The most popular method to extend the size of database in previous works is to resize the images into patches. In Jun 8, 2017 · Memory performance is a key bottleneck for deep learning systems. In this paper, we utilize and analyze the binarized neural network in doing human detection on infrared images. Neural networks has been successfully used in the processing of Lidar data, especially in the scenario of autonomous driving. DFF-Net includes two modules: the prior object-detection module and the object-detection module. Image classification and object detection are some of the oldest problems in computer vision that researchers have tried to solve for many decades. The convolutional neural network is trained with the images and bounding box annotation files of the transparent object classes. from torchvision. Sep 17, 2016 · Abstract. The speed-up is realized without sacrificing the object detection accuracy (mAP). 1. From the current academic research and practical application, the object detection algorithm based on the convolutional neural network has higher accuracy and shorter test time than the traditional method, and it has almost completely replaced the traditional algorithm. The results show that the application of deep learning method in detection of transparent objects can successfully perform the detection of transparent Jul 21, 2022 · Abstract: Automatic detection of magnetic tile surface defects is an important and challenging task, especially for small objects such as cracks. Specifically, we propose an end-to-end object detection network for reducing the impact of Feb 15, 2024 · In conclusion, the SSD neural network represents a significant milestone in the field of object detection, offering a sophisticated blend of speed, accuracy, and computational efficiency. R-CNN is an object detection framework, which uses a convolutional neural network (CNN) to classify image regions within an image [1]. 2. This paper proposed a knowledge graph network based on a graph convolution network to improve the accuracy of baseline detectors. Anything not binary is then object classification or object recognition (e. Binarization of both activations and weights is one promising approach that can best scale to realize the highest energy efficient system using the lowest possible precision. Use object detection when images contain multiple objects of different types. Using neural networks and deep learning, we have reached a stage where The convolutional neural networks (CNNs) represent the heart of state-of-the-art object detection methods. In the proposal sub-network, detection is performed at multiple output layers, so that receptive fields match objects of different scales. Today, many applications use object-detection networks as one of their main components. Consequently, we present a first spiked-based object detection model, called Spiking-YOLO. 3 Faster region convolutional neural network. See Deep learning vs machine learning for more information. 3389/fonc. The Mask Region-based Convolutional Neural Network, or Mask R-CNN, model is one of the state-of-the-art approaches for object recognition tasks. Several CNNs are available, for instance, AlexNet, VGGNet, and ResNet. The network was evaluated on the SUN RGB-D dataset for object detection and we obtained a positive difference in mAP score of about 4% May 4, 2023 · The neural network for object detection, in addition to the object type and probability, returns the coordinates of the object on the image: x, y, width and height, as shown on the second image. Our model significantly outperforms comparable Jan 4, 2024 · Region-based convolutional neural networks or regions with CNN features (R-CNNs) are pioneering approaches that apply deep models to object detection. The Open Neural Network Exchange (ONNX) is an open standard format created to represent machine learning models. Then we focus on typical generic object detection architectures along with some modifications and useful tricks Jul 15, 2019 · 3. While the deep learning-based image deraining methods have made great progress in recent years, there are two major shortcomings in their application in real-world situations. Oncol. Deep Object Detection Pipelines Most state-of-the-art deep neural network based object detection models are derived from multiple stages and sin-gle stage pipelines, starting from R-CNN [4] and YOLO [15], respectively. doi: 10. The prior module produces initial anchor boxes for the subsequent detection module. With images, the term "recognition" is often a better fit since there is always some uncertainty involved and "recognition" reflects those specific Aug 2, 2021 · In this section, you will learn how to perform object detection with pre-trained PyTorch networks. More specifically, a novel opto-electronic hybrid deep neural network that cascades an optical encoder, convolutional neural network (CNN) decoder and video object detection module to allow for end-to-end optimization is built for this task. They are used for extracting features. Although many traditional detection models have demonstrated an acceptable performance by using the imagery pyramid and multiple templates in a sliding-window manner, such techniques are inefficient and costly. Aug 21, 2021 · Region-based convolutional neural networks use a set of region proposals to detect objects. A unified deep neural network, denoted the multi-scale CNN (MS-CNN), is proposed for fast multi-scale object detection. An object detection model is trained to detect the presence and location of multiple classes of objects. Deep Neural Networks (DNNs) have recently shown outstanding performance on image classification tasks [14]. We train it to recognize only one class that encapsulates the general features of what it is we want to classify — a cat, a mobile app, a car brand logo. Deep Learning with OpenCV DNN Module: A Definitive Guide. Our review begins with a brief introduction on the history of deep learning and its representative tool, namely, the convolutional neural network. Our results show comparable algorithmic Jun 30, 2016 · Object Classification with CNNs Using the Keras Deep Learning Library. The deep CNNs are capable to learn prominent-feature representations of images due to their typical hierarchical architecture, and hence, it offers a fast, rapid, and accurate way to predict the position of objects within the image. Indeed, SNNs are more biologically realistic neural networks where neurons communicate using discrete and Oct 1, 2020 · Object detection is an essential function for mobile robots, allowing them to carry out missions efficiently. To solve this problem, we propose a novel convolutional neural network to accurately detect salient objects, by digging effective edge information from light-field data. These tweaks do not change the model architectures Click here to download the full example code. It seems that deep learning based approaches are currently more superior to the more "traditional" methods in the domain of object detection. NVIDIA GeForce RTX™ powers the world’s fastest GPUs and the ultimate platform for gamers and creators. py script and insert the following code: # import the necessary packages. Some use cases for object detection include: Self The drone video objection detection is challenging owing to the appearance deterioration, object occlusion and motion blur in video frames, which are caused by the object motion, the camera motion, and the mixture of the object motion and the camera motion in the drone video. One use of such a scenario is automatic photo organization in social media platforms. Compared with traditional handcrafted feature-based methods, the deep learning-based object detection methods can learn both low-level and high-level image features. Two-shot object detection uses two passes of the input image to make predictions about the presence and location of objects. These models behave differently in network architecture, training strategy, and optimization function. Feb 23, 2022 · An End-to-End Cascaded Image Deraining and Object Detection Neural Network. Object detection neural networks can also detect several objects in the image and their bounding boxes. Gaming and Creating. array tuple samples in the form: Mar 8, 2021 · Keywords: medical images, convolutional neural network, object detection, semantic segmentation, analysis. to object detection training to a certain degree. The Matterport Mask R-CNN project provides a library that […] Jun 12, 2020 · With the rapid development of deep learning techniques, deep convolutional neural networks (DCNNs) have become more important for object detection. Citation: Yang R and Yu Y (2021) Artificial Convolutional Neural Network in Object Detection and Semantic Segmentation for Medical Imaging Analysis. In particular, our method is May 9, 2023 · Object detection both locates and categorizes entities within images. Latest state-of-the-art detec-tors, including single (SSD [12] and YOLO [16]) or mul-tiple stage RCNN-like [17] neural networks, are based on Object detection is a challenging computer vision task that involves predicting both where the objects are in the image and what type of objects were detected. In single stage pipelines, predictions are generated by a single convolutional network and therefore Jul 31, 2015 · Object detection answers the question "Is the object detected?" (Yes/No). In this chapter, we have reviewed some popular state-of-the-art object detection models based on CNN. Neurosci. Open the detect_image. 3389/fnins. [ 10] proposed R-CNN that is a first region-based CNN detector as shown in Fig. Significant advances in object detection have been achieved through improved object representation and the use of deep neural network models. This capability is crucial in Apr 9, 2021 · In this paper, we report an optical and digital co-design architecture for video object detection from a single coded image (VODS). Besides the traditional object detection techniques, advanced deep learning models like R-CNN and YOLO can achieve impressive detection over different types of YOLO is a single-shot detector that uses a fully convolutional neural network (CNN) to process an image. To detect these objects in railways, we proposed an object-detection method using a differential feature fusion convolutional neural network (DFF-Net). Jul 21, 2023 · Spiking neural networks (SNNs) are brain-inspired energy-efficient models that encode information in spatiotemporal dynamics. Enjoy beautiful ray tracing, AI-powered DLSS, and much more in games and applications, on your desktop, laptop, in the cloud, or in your living room. import numpy as np. Instead of classifying every region using a sliding window, the R-CNN detector only processes those regions that are likely to contain an object. In computer vision, object detection is the problem of locating one or more objects in an image. , offsets). However, existing methods heavily rely on pre-processing of the pulse signals derived from Lidar sensors and therefore Feb 22, 2022 · In this paper, we explore the combination of the low-level vision task with the high-level vision task. In this works, we explore training tweaks that apply to various models including Faster R-CNN and YOLOv3. Feb 2, 2024 · Spiking Neural Networks (SNNs) are a promising approach to address this challenge, with their event-driven information flow and sparse activations. ·. Jan 1, 2023 · Object detection algorithms such as Faster RCNN [20], [21], and YOLO [23] are inadequate at detecting small size objects. The field of computer vision has existed since the late 1960s. This section describes the signature for Single-Shot Detector models converted to TensorFlow Lite from the TensorFlow Object Detection API. It combines an SNN CenterNet adaptation with an efficient M2U-Net-based decoder. Since the pooling operation causes information loss in the standard CNN, we designed a wavelet-multiresolution-analysis-based 3D object detection network without a pooling operation. The faster region convolutional neural network is another state-of-the-art CNN-based deep learning object detection approach. In this paper, we propose to use Deep Convolutional Neural Network (DCNN) features from combined layers to perform orientation robust aerial object detection. Sep 11, 2017. Using the trained network, transparent objects are detected from the images. TorchVision Object Detection Finetuning Tutorial ¶. Recently, convolutional neural networks (CNNs) have Deep architectures for object detection and parsing have been motivated by part-based models and traditionally are called compositional models, where the object is expressed as layered composition of image primitives. The general trend in object detection has been to design large and over-parameterized models, which can achieve excellent performance. One of the typical solutions is to use Convolutional Neural Networks (CNNs) to train detection model by taking single Evaluating Deep Learning Models: The Confusion Matrix, Accuracy, Precision, and Recall. 30. Apr 24, 2022 · FOMO is a deep learning object detection model that weighs less than 200 kilobytes. This paper presents a new convolutional neural network architecture for performing object detection based on RGB-D images. Convolutional neural network (CNN) has turned to be the state of the art for object detection task of computer vision. We Mar 12, 2019 · Through our in-depth analysis, we introduce two novel methods: channel-wise normalization and signed neuron with imbalanced threshold, both of which provide fast and accurate information transmission for deep SNNs. The fast region-based convolutional neural network (Fast R-CNN) was proposed by Ross Girshick in 2015 and was designed to improve the speed and accuracy of object detection. The article discusses the integration of the YOLOv3 neural network on the DE10-Nano FPGA. Introduction Object detection is no doubt one of the cutting edge applications in computer vision drawing attentions of re-searchers from various fields. Subsequently, an object detection method based on dynamic neural network is proposed. Our GPU-accelerated faster R-CNN can detect objects with 55. The best-of-breed open source library implementation of the YOLOv3 for the Keras deep learning library. Popular deep learning–based approaches using convolutional neural networks (CNNs), such as R-CNN and YOLO v2, automatically learn to detect objects within images. A difficult problem where traditional neural networks fall down is called object recognition. Convolutional Neural Network. This computer vision task has a wide range of applications, from medical imaging to self-driving cars. This one we train — hopefully — only once. The performance of both networks is evaluated on a real world dataset with dense city traffic scenarios, diverse object orientations and distances as well as occlusions up to Jun 20, 2023 · Object detection is one of the most basic and important research tasks in the field of computer vision. Among the Neural Network based approaches — we can classify the algorithms in two main categories — single stage and two stage object detector. This is nearly 30% speed-up. Our review begins with a brief introduction on the history of deep learning and its representative tool, namely Convolutional Neural Network (CNN). models import detection. However, the existing solution for classification of radar echo signal is limited because its deterministic analysis is too complicated to describe various object features. 11:638182. Two-shot object detection. **Object Detection** is a computer vision task in which the goal is to detect and locate objects of interest in an image or video. We explore the inherent characteristics of DC-NN as well as relate the Jul 23, 2022 · Convolutional neural networks (CNNs) have been widely applied in the image quality assessment (IQA) field, but the size of the IQA databases severely limits the performance of the CNN-based IQA models. It needs a more Jun 21, 2021 · In the past few years, deep learning object detection has come a long way, evolving from a patchwork of different components to a single neural network that works efficiently. We also design the components of the loss function to accommodate the characteristics of the different sub-networks. In YOLO a single convolutional network predicts the bounding boxes and the class Jul 1, 2021 · Object detection is emerging as a subdomain of computer vision (CV) that benefits from DL, especially convolutional neural networks (CNNs) [7]. 150. The data set is a Python dict of the form: {‘samples’: samples, ‘labels’: labels} samples is a list of N radar projection numpy. We propose Spiking CenterNet for object detection on event data. CNN is an improved version of multilayer Aug 30, 2023 · Model description. Methods like YOLO, for example, seem to be doing something magical that can't be replicated by the other methods. Supported by a robust community of partners, ONNX defines a common set of operators and a common file format to enable AI developers to use models with a variety of frameworks, tools, runtimes, and compilers. 2. For example, a model might be trained with images that contain various pieces of Specifically, we propose an end-to-end object detection network for reducing the impact of rainfall, which consists of two cascaded networks, an improved image deraining network and an object detection network, respectively. Object detection models are commonly trained using deep learning and neural networks. g. Object detection is a technique that uses neural networks to localize and classify objects in images. YOLO or You Only Look Once is an object detection algorithm much different from the region based algorithms seen above. Initially, an attention region estimation method is proposed. These networks are mainly used for object classification task and have evaluated on some widely used benchmarks and datasets such Dec 27, 2021 · Convolutional neural networks have pushed forward image analysis research and computer vision over the last decade, constituting a state-of-the-art approach in object detection today. However, human visual system (HVS) can only perceive the qualities of objects in an Detecting objects in aerial images is a challenging task due to multiple orientations and relatively small size of the objects. Dec 10, 2015 · Detecting objects in aerial images is challenged by variance of object colors, aspect ratios, cluttered backgrounds, and in particular, undetermined orientations. This method conceptualizes the estimation of the attention region as a mass center computation, taking into account the ego-vehicle’s driving state and the surrounding traffic participants. In this work, we propose to train spiking neural networks (SNNs) directly on data coming from event cameras to design fast and efficient automotive embedded applications. Jun 7, 2018 · In this paper, we apply one of the Convolutional Neural Network called Single Shot MultiBox Detector (SSD) for transparent object detection task and evaluate the performance of the system. This article Sep 11, 2017 · 11 min read. In deep convolutional neural network, there is a lack of adequate knowledge in independent feature layers as they occupy only a small pixel size in the actual image. Firstly, the gap between the low-level vision task represented by rain removal and the high Jun 6, 2021 · This project employs autonomous supervised learning whereby standard camera-based object detection techniques are used to automatically label radar scans of people and objects. Our experiments show that Spiking-YOLO Dec 3, 2021 · Considering this problem in terms of faulty small-object detection, this study proposed an improved faster regional convolutional neural network-based model to detect the faults in the product images. You can choose from two key approaches to get started with object detection using deep learning: Jun 9, 2020 · Abstract. Jul 5, 2024 · The domain of object detection was revolutionized with the introduction of Convolutional Neural Networks (CNNs) in the field of computer vision. the following video shows a FOMO neural network detecting objects at 30 frames per second on an Arduino Nov 18, 2017 · The Region-Convolutional Neural Network (R-CNN) A few years ago, by exploiting some of the leaps made possible in computer vision via CNNs, researchers developed R-CNNs to deal with the tasks of object detection, localization and classification. Paper. “A convolutional neural network is a class of deep neural networks, most commonly applied to analyzing visual imagery. Object detection: CNNs are able to identify and locate multiple objects within an image. It is where a model is able to identify the objects in images. ). Recently, deep SNNs trained directly have shown great success in achieving high performance on classification tasks with very few time steps. The design of increasingly deeper and wider architectures has made it possible to achieve unprecedented levels of detection accuracy, albeit at the cost of both a dramatic computational burden and a large memory Feb 23, 2024 · Literature 7 coupled the physical lighting model with a deep neural network to improve image quality Huang, M. However, this comes at the expense of low speed, heavy computation and large amount of memory overhead, also makes object detection models more difficult to Oct 29, 2018 · Object Detection based on LIDAR Temporal Pulses using Spiking Neural Networks. In recent years, various deep learning models based on convolutional neural networks However, a noteworthy effort has been carried out in the past years in the detection of objects using convolutional neural networks (CNNs). This article aims to explore the architectural intricacies, methodological differences, and performance characteristics of three CNN-based object detection algorithms, namely Faster Region-Based Convolutional Network (R-CNN), You Only Look Once v3 Radar-based object detection becomes a more important problem as such sensor technology is broadly adopted in many applications including military, robotics, space exploring, and autonomous vehicles. D. Mar 28, 2022 · The detection result of current light-field salient object detection methods suffers from loss of edge details, which significantly limits the performance of subsequent computer vision tasks. We will dive deeper into the YOLO model in the next section. Additionally, instead of using a single filter like the standard convolution, we used the lower-frequency You can use a variety of techniques to perform object detection. This greatly reduces the computational cost incurred when running Detecting objects remains one of computer vision and image understanding applications’ most fundamental and challenging aspects. Girshick et al. 2021. R-CNN models first select several proposed regions from an image (for example, anchor boxes are one type of selection method) and then label their categories and bounding boxes (e. Object detection is one of the most active research and application areas of neural networks. Slightly worse indicators of the main metrics (mAP, FPS, inference time) when operating a neural Image classification: Convolutional neural networks are used for image categorization, where images are assigned to predefined categories. In this article, a spatial scale adaptive real-time object detection neural network, namely YoloOW, was proposed to tackle the challenge of personnel and boat detection amid the diverse UAV imagery, which comprises a feature extractor, a feature enhancer, and a postprocessor. A notable example is the And=Or graph [20], where an object is modeled by a tree with And-nodes representing different parts and Jul 15, 2018 · In this paper, we provide a review on deep learning based object detection frameworks. Most of the recent successful object detection methods are based on convolutional neural networks (CNNs). Citation: van Dyck LE, Kwitt R, Denzler SJ and Gruber WR (2021) Comparing Object Recognition in Humans and Deep Convolutional Neural Networks—An Eye Tracking Study. In detection networks, the processes of building scored candidate regions, sorting and non-maximum-suppression (nms) are commonly used. In this architecture, the network takes the provided input image into a convolutional network which provides a convolutional feature map. It is a successor to R 2. Several techniques for object detection using deep learning are available such as Faster R-CNN, you only look once (YOLO) v2, YOLO v3, YOLO v4, YOLOX, and single shot detection (SSD). To address this problem, we propose a novel framework called feature enhancement and loop-shaped fusion convolutional neural network (ELCNN), which enhances shallow features and fuses features with a loop-shaped feature pyramid structure. We have made a categorization of those detection models according to two different approaches: two-stage approach Feb 11, 2019 · Object detection models, however, have more complex neural network structures and optimization targets. DENet: Detection-driven enhancement network for object detection under Nov 20, 2023 · Object detection (OD) is growing rapidly due to the rebirth of convolution neural networks. In this paper, we provide a review of deep learning-based object detection frameworks. Shibo Zhou, Wei Wang. oa qd ep xe bm qh ds ae pl wu