Ming Wu

Dr. Ming Wu


Research Focus & Metrics

Academic Career Length

17Years (2010-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 February 2020 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
DCDongliang Chang
Dongliang Chang
YDYifeng Ding
Yifeng Ding
JXJiyang Xie
Jiyang Xie
ABAyan Kumar Bhunia
Ayan Kumar Bhunia
XLXiaoxu Li
Xiaoxu Li
YSYi-Zhe Song
Yi-Zhe Song
The key to solving fine-grained image categorization is finding discriminate and local regions that correspond to subtle visual traits. Great strides have been made, with complex networks designed specifically to learn part-level discriminate feature representations. In this paper, we show that it is possible to cultivate subtle details without the need for overly complicated network designs or training mechanisms – a single loss is all it takes. The main trick lies with how we delve into individual feature channels early on, as opposed to the convention of starting from a consolidated feature map. The proposed loss function, termed as mutual-channel loss (MC-Loss), consists of two channel-specific components: a discriminality component and a diversity component. The discriminality component forces all feature channels belonging to the same class to be discriminative, through a novel channel-wise attention mechanism. The diversity component additionally constraints channels so that they become mutually exclusive across the spatial dimension. The end result is therefore a set of feature channels, each of which reflects different locally discriminative regions for a specific class. The MC-Loss can be trained end-to-end, without the need for any bounding-box/part annotations, and yields highly discriminative regions during inference. Experimental results show our MC-Loss when implemented on top of common base networks can achieve state-of-the-art performance on all four fine-grained categorization datasets (CUB-Birds, FGVC-Aircraft, Flowers-102, and Stanford Cars). Ablative studies further demonstrate the superiority of the MC-Loss when compared with other recently proposed general-purpose losses for visual classification, on two different base networks.

8 February 2021 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
YDYifeng Ding
Yifeng Ding
ZMZhanyu Ma
Zhanyu Ma
SWShaoguo Wen
Shaoguo Wen
JXJiyang Xie
Jiyang Xie
DCDongliang Chang
Dongliang Chang
HLHaibin Ling
Haibin Ling
Classifying the sub-categories of an object from the same super-category (e.g., bird species and cars) in fine-grained visual classification (FGVC) highly relies on discriminative feature representation and accurate region localization. Existing approaches mainly focus on distilling information from high-level features. In this article, by contrast, we show that by integrating low-level information (e.g., color, edge junctions, texture patterns), performance can be improved with enhanced feature representation and accurately located discriminative regions. Our solution, named Attention Pyramid Convolutional Neural Network (AP-CNN), consists of 1) a dual pathway hierarchy structure with a top-down feature pathway and a bottom-up attention pathway, hence learning both high-level semantic and low-level detailed feature representation, and 2) an ROI-guided refinement strategy with ROI-guided dropblock and ROI-guided zoom-in operation, which refines features with discriminative local regions enhanced and background noises eliminated. The proposed AP-CNN can be trained end-to-end, without the need of any additional bounding box/part annotation. Extensive experiments on three popularly tested FGVC datasets (CUB-200-2011, Stanford Cars, and FGVC-Aircraft) demonstrate that our approach achieves state-of-the-art performance. Models and code are available at https://github.com/PRIS-CV/AP-CNN_Pytorch-master.

26 April 2019 in IEEE Access
Verified
Journal Article
Open access
MWMing Wu
Ming Wu
CZChuang Zhang
Chuang Zhang
JLJiaming Liu
Jiaming Liu
LZLichen Zhou
Lichen Zhou
XLXiaoqi Li
Xiaoqi Li
Satellite image semantic segmentation including extracting road, detecting building, and identifying land cover types are essential for sustainable development, agriculture, forestry, urban planning and climate change research. Nevertheless, it is still unclear how to develop a refined semantic segmentation model in an efficient and elegant way. In this paper, we propose AD-LinkNet(Attention Dilation-LinkNet) neural network, which adopts encoder-decoder structure, serial-parallel combination dilated convolution, channel-wise attention mechanism and pretrained encoder for semantic segmentation. Serial-parallel combination dilated convolution enlarges receptive field as well as assemble multi-scale features for multi-scale objects, such as long span road and small pool. Channel-wise attention mechanism are designed to advantage the context information in satellite image. Experimental results on road extraction and surface classification data sets prove that AD-LinkNet shows a significant effect on improving segmentation accuracy. We defeated the D-Linknet algorithm that won the first place in the CVPR 2018 DeepGlobe road extraction competition.

3 September 2022 in Remote Sensing
Verified
Journal Article
Open access
PDF file
MXMengqiu XU
Mengqiu XU
MWMing Wu
Ming Wu
KCKaixin Chen
Kaixin Chen
CZChuang Zhang
Chuang Zhang
JGJun Guo
Jun Guo
With the rapid development of the remote sensing monitoring and computer vision technology, the deep learning method has made a great progress to achieve applications such as earth observation, climate change and even space exploration. However, the model trained on existing data cannot be directly used to handle the new remote sensing data, and labeling the new data is also time-consuming and labor-intensive. Unsupervised Domain Adaptation (UDA) is one of the solutions to the aforementioned problems of labeled data defined as the source domain and unlabeled data as the target domain, i.e., its essential purpose is to obtain a well-trained model and tackle the problem of data distribution discrepancy defined as the domain shift between the source and target domain. There are a lot of reviews that have elaborated on UDA methods based on natural data, but few of these studies take into consideration thorough remote sensing applications and contributions. Thus, in this paper, in order to explore the further progress and development of UDA methods in remote sensing, based on the analysis of the causes of domain shift, a comprehensive review is provided with a fine-grained taxonomy of UDA methods applied for remote sensing data, which includes Generative training, Adversarial training, Self-training and Hybrid training methods, to better assist scholars in understanding remote sensing data and further advance the development of methods. Moreover, remote sensing applications are introduced by a thorough dataset analysis. Meanwhile, we sort out definitions and methodology introductions of partial, open-set and multi-domain UDA, which are more pertinent to real-world remote sensing applications. We can draw the conclusion that UDA methods in the field of remote sensing data are carried out later than those applied in natural images, and due to the domain gap caused by appearance differences, most of methods focus on how to use generative training (GT) methods to improve the model’s performance. Finally, we describe the potential deficiencies and further in-depth insights of UDA in the field of remote sensing.


5 June 2019 in IEEE Access
Verified
Journal Article
Open access
CZChuang Zhang
Chuang Zhang
MWMing Wu
Ming Wu
JCJinyu Chen
Jinyu Chen
KCKaiyan Chen
Kaiyan Chen
CZChi Zhang
Chi Zhang
ZHZichen He
Zichen He
Visibility affects all forms of traffic: roads, sailing and aviation. Visibility prediction is meaningful in guiding production and life. Different from weather prediction, which relies solely on atmosphere factors, the factors that affect the meteorogical visibility are more complicated, such as the air pollution caused by factory exhaust emission. However, the current prediction of visibility is mostly based on the numerical prediction method similar to the weather prediction. We proposed a method using multimodal fusion to build a weather visibility prediction system in this paper. An advanced numerical prediction model and a method for emission detection were used to build a multimodal fusion visibility prediction system. We used the most advanced regression algorithm, XGBoost and LightGBM, to train the fusion model for numerical prediction. Through estimation of factory emission by the traditional detector in the satellite image, we propose to add the result of estimation based on Landsat-8 satellite images to assist the prediction. By testing our numerical model in atmosphere data of various meteorological observation stations in Beijing-Tianjin-Hebei region from 2002 to 2018, our numerical prediction model turns out to be more accuracy than other existing methods, and after fusing with emission detection method, the accuracy of our visibility prediction system has been further improved.