检索结果-内蒙古大学图书馆

Discriminative Asymmetric Learning for Efficient Surgical Instrument Parsing

学校读者我要写书评

暂无评论

Discriminative Asymmetric Learning for Efficient Surgical In...

IEEE International Conference on Robotics and Automation (ICRA)

作者： Jiaqi Liu Yu Qiao Jie Yang Guang-Zhong Yang Yun Gu Institute of Medical Robotics Shanghai Jiao Tong University Shanghai China Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China

Semantic segmentation of surgical instruments provides essential priors for autonomous surgery. This task is however challenging since the fine-structure of surgical instruments requires the accurate segmentation of detailed regions in images. As the visual guidance for autonomous surgery, the algorithm should also be real-time and friendly to embedded systems. In this paper, a discriminative asymmetric learning framework is proposed to balance the efficiency and effectiveness of surgical instrument segmentation. Two convolutional neural networks with specific designs are deployed to extract the detail and semantic features of instruments. To reduce the redundancy of visual representation, the aggregator-discriminator mechanism is proposed to distinguish the features learned from different levels. Experiments demonstrate that the proposed method contributes to competitive segmentation accuracy and a higher efficiency compared to existing methods.

关键词： image segmentation Visualization Embedded systems Instruments Semantics Redundancy Surgery

Learning with asymmetric kernels: Least squares and feature interpretation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： He, Mingzhen He, Fan Shi, Lei Huang, Xiaolin Suykens, Johan A.K. Institute Of Image Processing And Pattern Recognition Shanghai Jiao Tong University Shanghai200240 China Shanghai Key Laboratory For Contemporary Applied Mathematics Fudan University Shanghai200433 China School Of Mathematical Sciences Fudan University Shanghai200433 China KU Leuven LeuvenB-3001 Belgium

Asymmetric kernels naturally exist in real life, e.g., for conditional probability and directed graphs. However, most of the existing kernel-based learning methods require kernels to be symmetric, which prevents the use of asymmetric kernels. This paper addresses the asymmetric kernel-based learning in the framework of the least squares support vector machine named AsK-LS, resulting in the first classification method that can utilize asymmetric kernels directly. We will show that AsK-LS can learn with asymmetric features, namely source and target features, while the kernel trick remains applicable, i.e., the source and target features exist but are not necessarily known. Besides, the computational burden of AsK-LS is as cheap as dealing with symmetric kernels. Experimental results on the Corel database, directed graphs, and the UCI database will show that, in the case asymmetric information is crucial, the proposed AsK-LS can learn with asymmetric kernels and performs much better than the existing kernel methods that have to do symmetrization to accommodate asymmetric kernels. © 2022, CC BY.

关键词： Directed graphs

Riemannian Self-Attention Mechanism for SPD Networks

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wang, Rui Wu, Xiao-Jun Li, Hui Kittler, Josef School of Artificial Intelligence and Computer Science Jiangnan University Wuxi214122 China Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence Jiangnan University China Centre for Vision Speech and Signal Processing University of Surrey GuildfordGU2 7XH United Kingdom

Symmetric positive definite (SPD) matrix has been demonstrated to be an effective feature descriptor in many scientific areas, as it can encode spatiotemporal statistics of the data adequately on a curved Riemannian manifold, i.e., SPD manifold. Although there are many different ways to design network architectures for SPD matrix nonlinear learning, very few solutions explicitly mine the geometrical dependencies of features at different layers. Motivated by the great success of self-attention mechanism in capturing long-range relationships, an SPD manifold self-attention mechanism (SMSA) is proposed in this paper using some manifold-valued geometric operations, mainly the Riemannian metric, Riemannian mean, and Riemannian optimization. Then, an SMSA-based geometric learning module (SMSA-GLM) is designed for the sake of improving the discrimination of the generated deep structured representations. Extensive experimental results achieved on three benchmarking datasets show that our modification against the baseline network further alleviates the information degradation problem and leads to improved accuracy. Copyright © 2023, The Authors. All rights reserved.

关键词： Network architecture

Semi-Supervised Skin Lesion Segmentation with Learning Model Confidence

学校读者我要写书评

暂无评论

Semi-Supervised Skin Lesion Segmentation with Learning Model...

IEEE International Conference on Acoustics, Speech and Signal processing

作者： Zhiqiang Xie Enmei Tu Hao Zheng Yun Gu Jie Yang Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China Institute of Medical Robotics Shanghai Jiao Tong University Shanghai China

ISBN: (纸本)9781728176055;9781728176062

Segmentation of skin lesions is important for disease diagnoses and treatment planning. Over the years, semi-supervised methods using pseudo labels have boosted the segmentation performance with limited labeled data and abundant unlabeled data. However, the unreliable targets in pseudo labels might lead to meaningless guidance for unlabeled data. In this paper, to solve this issue, we propose a novel confidence aware semi-supervised learning method based on a mean teacher scheme. Concretely, we design a confidence module to predict the model confidence guided by the True Class Probability. Then in the mean teacher framework, the student model gradually learns trustworthy targets from teacher model. To further improve the segmentation quality, we fine-tune the student model with reliable content in pseudo labels. We conduct extensive experiments on 2018 ISIC skin lesion segmentation dataset and our method outperforms other state-of-the-art semi-supervised approaches.

关键词： Conferences Semisupervised learning Signal processing Predictive models Skin Planning Lesions

Transfer Learning of Fuzzy Spatio-Temporal Rules in the NeuCube Brain-Inspired Spiking Neural Network: A Case Study on EEG Spatio-temporal Data

学校读者我要写书评

暂无评论

TechRxiv

TechRxiv 2022年

作者： Kasabov, Nikola K. Tan, Yongyao Doborjeh, Maryam Tu, Enmei Yang, Jie School of Engineering Computer and Mathematical Science KEDRI Auckland University of Technology New Zealand Intelligent Systems Research Centre University of Ulster United Kingdom Bulgarian Academy of Sciences Bulgaria Institute of Image Processing and Pattern Recognition Shanghai Jiao Tiong University China

The paper demonstrates for the first time that a brain-inspired spiking neural network (SNN) architecture can be used not only to learn spatio-temporal data, but also to extract fuzzy spatio-temporal rules from such data and to update these rules incrementally in a transfer learning mode. We propose a method, where a SNN model learns incrementally new time-space data related to new classes/tasks/categories, always utilising some previously learned knowledge, and presents the evolved knowledge as fuzzy spatio-temporal rules. Similarly, to how the brain manifests transfer learning, these SNN models do not need to be restricted in number of layers, neurons in each layer, etc. as they adopt self-organising learning principles. The continuously evolved fuzzy rules from spatio-temporal data are interpretable for a better understanding of the processes that generate the data. The proposed method is based on a brain-inspired SNN architecture NeuCube and is illustrated on tasks of incremental and transfer learning and knowledge transfer using spatio-temporal EEG data measuring brain activity, when persons performing tasks in space and time. The method is a general one and opens the field to create new types of adaptable and explainable AI systems for applications across domain areas, e.g. brain-machine interfaces, robotics, speech and sound processing, cognitive studies and the creation of lifelong-learning cognitive machines. © 2022, CC BY.

关键词： Brain

Depth Prior-Based Stable Tensor Decomposition for Video Snow Removal

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Li, Yongji Wang, Luping Jia, Zhenhong Yang, Jie Kasabov, Nikola School of Electronics and Communication Engineering Sun Yat-sen University Guangzhou510275 China College of Information Science and Engineering Xinjiang University Urumqi830046 China Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai200400 China Knowledge Engineering and Discovery Research Institute Auckland University of Technology Auckland1020 New Zealand

Heavy snow seriously degrades the performance of outdoor computer vision systems. Near- and far-field snowflakes in heavy snow videos exhibit distinctly disparate physical properties. To address this issue, this research proposes a video desnowing model that utilizes stable tensor decomposition with snow depth prior information. Initially, the depth of snowflakes is transformed by their speeds calculated from the optical flow field of snowfall. Next, the noise level in backgrounds is available by dense snow. In contrast to previous tensor decomposition (TD), the introduction of a noise term enhances the stability of the depth prior-based stable tensor decomposition (DP-STD) in addressing heavy snowfall. This leads to more efficient and accurate restoration of underlying nonsnow structure. Inspired by biomimicry, the snow on moving objects (MOs) is removed by the adaptive mimesis region of interest (AM-ROI). Both synthetic and real snowfall experimental results show that our proposed desnowing model is more effective than the current SOTA algorithm in removing heavy snow. © 2023, The Authors. All rights reserved.

关键词： Snow

Fdflownet: Fast Optical Flow Estimation Using A Deep Lightweight Network

学校读者我要写书评

暂无评论

Fdflownet: Fast Optical Flow Estimation Using A Deep Lightwe...

IEEE International Conference on image processing

作者： Lingtong Kong Jie Yang Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University China

ISBN: (数字)9781728163956

ISBN: (纸本)9781728163963

Significant progress has been made for estimating optical flow using deep neural networks. Advanced deep models achieve accurate flow estimation often with a considerable computation complexity and time-consuming training processes. In this work, we present a lightweight yet effective model for real-time optical flow estimation, termed FDFlowNet (fast deep flownet). We achieve better or similar accuracy on the challenging KITTI and Sintel benchmarks while being about 2 times faster than PWC-Net. This is achieved by a carefully-designed structure and newly proposed components. We first introduce an U-shape network for constructing multi-scale feature which benefits upper levels with global receptive field compared with pyramid network. In each scale, a partial fully connected structure with dilated convolution is proposed for flow estimation that obtains a good balance among speed, accuracy and number of parameters compared with sequential connected and dense connected structures. Experiments demonstrate that our model achieves state-of-the-art performance while being fast and lightweight.

关键词： Estimation Optical imaging Convolution Adaptive optics Optical fiber networks Computational modeling Training

Sequence-level training for non-autoregressive neural machine translation

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Shao, Chenze Feng, Yang Zhang, Jinchao Meng, Fandong Zhou, Jie Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Pattern Recognition Center WeChat AI Tencent Inc

In recent years, Neural Machine Translation (NMT) has achieved notable results in various translation tasks. However, the word-by-word generation manner determined by the autoregressive mechanism leads to high translation latency of the NMT and restricts its low-latency applications. Non-Autoregressive Neural Machine Translation (NAT) removes the autoregressive mechanism and achieves significant decoding speedup through generating target words independently and simultaneously. Nevertheless, NAT still takes the word-level cross-entropy loss as the training objective, which is not optimal because the output of NAT cannot be properly evaluated due to the multimodality problem. In this article, we propose using sequence-level training objectives to train NAT models, which evaluate the NAT outputs as a whole and correlates well with the real translation quality. Firstly, we propose training NAT models to optimize sequence-level evaluation metrics (e.g., BLEU) based on several novel reinforcement algorithms customized for NAT, which outperforms the conventional method by reducing the variance of gradient estimation. Secondly, we introduce a novel training objective for NAT models, which aims to minimize the Bag-of-Ngrams (BoN) difference between the model output and the reference sentence. The BoN training objective is differentiable and can be calculated efficiently without doing any approximations. Finally, we apply a three-stage training strategy to combine these two methods to train the NAT model. We validate our approach on four translation tasks (WMT14 En↔De, WMT16 En↔Ro), which shows that our approach largely outperforms NAT baselines and achieves remarkable performance on all translation tasks. The source code is available at https://***/ictnlp/Seq-NAT. Copyright © 2021, The Authors. All rights reserved.

关键词： Neural machine translation

Random Fourier Features for Asymmetric kernels

学校读者我要写书评

暂无评论

arXiv 2022年

作者： He, Mingzhen He, Fan Liu, Fanghui Huang, Xiaolin Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University 800 Dongchuan RD. Minhang District Shanghai200240 China Laboratory for Information and Inference Systems École Polytechnique Fédérale de Lausanne EPFL SV-BMC AA-B118 Lausanne1015 Switzerland

The random Fourier features (RFFs) method is a powerful and popular technique in kernel approximation for scalability of kernel methods. The theoretical foundation of RFFs is based on the Bochner theorem (Bochner, 2005) that relates symmetric, positive definite (PD) functions to probability measures. This condition naturally excludes asymmetric functions with a wide range applications in practice, e.g., directed graphs, conditional probability, and asymmetric kernels. Nevertheless, understanding asymmetric functions (kernels) and its scalability via RFFs is unclear both theoretically and empirically. In this paper, we introduce a complex measure with the real and imaginary parts corresponding to four finite positive measures, which expands the application scope of the Bochner theorem. By doing so, this framework allows for handling classical symmetric, PD kernels via one positive measure;symmetric, non-positive definite kernels via signed measures;and asymmetric kernels via complex measures, thereby unifying them into a general framework by RFFs, named AsK-RFFs. Such approximation scheme via complex measures enjoys theoretical guarantees in the perspective of the uniform convergence. In algorithmic implementation, to speed up the kernel approximation process, which is expensive due to the calculation of total mass, we employ a subset-based fast estimation method that optimizes total masses on a sub-training set, which enjoys computational efficiency in high dimensions. Our AsK-RFFs method is empirically validated on several typical large-scale datasets and achieves promising kernel approximation performance, which demonstrate the effectiveness of AsK-RFFs. © 2022, CC BY.

关键词： Computational efficiency