检索结果-内蒙古大学图书馆

Fast and accurate single-image depth estimation on mobile devices, mobile AI 2021 challenge: Report

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Ignatov, Andrey Malivenko, Grigory Plowman, David Shukla, Samarth Timofte, Radu Zhang, Ziyu Wang, Yicheng Huang, Zilong Luo, Guozhong Yu, Gang Fu, Bin Wang, Yiran Li, Xingyi Shi, Min Xian, Ke Cao, Zhiguo Du, Jin-Hua Wu, Pei-Lin Ge, Chao Yao, Jiaoyang Tu, Fangwen Li, Bo Yoo, Jung Eun Seo, Kwanggyoon Xu, Jialei Li, Zhenyu Liu, Xianming Jiang, Junjun Chen, Wei-Chi Joya, Shayan Fan, Huanhuan Kang, Zhaobing Li, Ang Feng, Tianpeng Liu, Yang Sheng, Chuannan Yin, Jian Benavides, Fausto T. Computer Vision Lab ETH Zurich Switzerland Ltd AI Witchlabs Switzerland Tencent GY-Lab China Key Laboratory of Image Processing and Intelligent Control Ministry of Education School of Artificial Intelligence and Automation Huazhong University of Science and Technology China Nanjing Artificial Intelligence Chip Research Institute of Automation Chinese Academy of Sciences China Black Sesame Technologies Inc. Singapore Singapore Visual Media Lab KAIST Korea Republic of Harbin Institute of Technology China Peng Cheng Laboratory China Multimedia and Computer Vision Laboratory National Cheng Kung University Taiwan Samsung Research UK United Kingdom OPPO Research Institute China ETH Zurich Switzerland

Depth estimation is an important computer vision problem with many practical applications to mobile devices. While many solutions have been proposed for this task, they are usually very computationally expensive and thus are not applicable for on-device inference. To address this problem, we introduce the first Mobile AI challenge, where the target is to develop an end-to-end deep learning-based depth estimation solutions that can demonstrate a nearly real-time performance on smartphones and IoT platforms. For this, the participants were provided with a new large-scale dataset containing RGB-depth image pairs obtained with a dedicated stereo ZED camera producing high-resolution depth maps for objects located at up to 50 meters. The runtime of all models was evaluated on the popular Raspberry Pi 4 platform with a mobile ARM-based Broadcom chipset. The proposed solutions can generate VGA resolution depth maps at up to 10 FPS on the Raspberry Pi 4 while achieving high fidelity results, and are compatible with any Android or Linux-based mobile devices. A detailed description of all models developed in the challenge is provided in this paper. Copyright © 2021, The Authors. All rights reserved.

关键词： computer operating systems

Effective fusion of deep multitasking representations for robust visual tracking

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Marvasti-Zadeh, Seyed Mojtaba Ghanei-Yakhdan, Hossein Kasaei, Shohreh Nasrollahi, Kamal Moeslund, Thomas B. Department of Electrical Engineering Yazd University Yazd Iran Sharif University of Technology Tehran Iran Vision and Learning Lab. University of Alberta Edmonton Canada Department of Electrical Engineering Yazd University Yazd Iran Department of Computer Engineering Sharif University of Technology Tehran Iran Department of Architecture Design and Media Technology Aalborg University Aalborg Denmark

Visual object tracking remains an active research field in computer vision due to persisting challenges with various problem-specific factors in real-world scenes. Many existing tracking methods based on discriminative correlation filters (DCFs) employ feature extraction networks (FENs) to model the target appearance during the learning process. However, using deep feature maps extracted from FENs based on different residual neural networks (ResNets) has not previously been investigated. This paper aims to evaluate the performance of twelve state-of-the-art ResNet-based FENs in a DCF-based framework to determine the best for visual tracking purposes. First, it ranks their best feature maps and explores the generalized adoption of the best ResNet-based FEN into another DCF-based method. Then, the proposed method extracts deep semantic information from a fully convolutional FEN and fuses it with the best ResNet-based feature maps to strengthen the target representation in the learning process of continuous convolution filters. Finally, it introduces a new and efficient semantic weighting method (using semantic segmentation feature maps on each video frame) to reduce the drift problem. Extensive experimental results on the well-known OTB-2013, OTB-2015, TC-128, and VOT-2018 visual tracking datasets demonstrate that the proposed method effectively outperforms state-of-the-art methods in terms of precision and robustness of visual tracking. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep neural networks

Flexible social inference facilitates targeted social learning when rewards are not observable

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Hawkins, Robert D. Berdahl, Andrew M. Pentland, Alex Tenenbaum, Joshua B. Goodman, Noah D. Krafft, P.M. Department of Psychology Stanford University StanfordCA United States Department of Psychology University of Wisconsin–Madison MadisonWI United States School of Aquatic and Fishery Sciences University of Washington SeattleWA United States Massachusetts Institute of Technology Media Lab. MIT CambridgeMA United States Department of Brain and Cognitive Sciences MIT CambridgeMA United States Department of Computer Science Stanford University StanfordCA United States Creative Computing Institute University of Arts London London United Kingdom

Groups coordinate more effectively when individuals are able to learn from others’ successes. But acquiring such knowledge is not always easy, especially in real-world environments where success is hidden from public view. We suggest that social inference capacities may help bridge this gap, allowing individuals to update their beliefs about others’ underlying knowledge and success from observable trajectories of behavior. We compared our social inference model against simpler heuristics in three studies of human behavior in a collective sensing task. In Experiment 1, we found that average performance improves as a function of group size at a rate greater than predicted by non-inferential models. Experiment 2 introduced artificial agents to evaluate how individuals selectively rely on social information. Experiment 3 generalized these findings to a more complex reward landscape. Taken together, our findings provide new insight into the relationship between individual social cognition and the flexibility of collective behavior. Copyright © 2022, The Authors. All rights reserved.

关键词： Behavioral research

Improvement of the Human Action Recognition Algorithm by the Pre-processing of Input Data

学校读者我要写书评

暂无评论

IOP Conference Series: Materials Science and Engineering 2021年第1期1029卷

作者： M M Zhdanova O S Balab.eva V V Voronin E A Semenishchev A A Zelensky Center for Cognitive Technology and Machine Vision Moscow State University of Technology «STANKIN» Moscow Russia Lab. «Mathematical methods of image processing and intelligent computer vision systems» Don State Technical University Rostov-on-Don Russia

The paper presents an approach to recognizing human actions using an additional preprocessing stage of input data. The growing volumes of video information do not always allow support the quality of data at a high level; this can cause limitations in the further processing of digital data. In this regard, it becomes urgent to introduce an additional stage of image enhancement into the algorithm for recognizing actions in video. The proposed method includes three main steps: image enhancement, constructing a descriptor, and classification. The presented image enhancement stage is based on the combined local and global image processing in the frequency domain. The basic idea in using local alfa-rooting method is to apply it to different disjoint blocks with different sizes. To solve the problem of constructing a descriptor, a three-dimensional microblock dense difference (3D DMD) algorithm is used, which provides a highly oriented representation of image regions by tightly capturing microblocks within each region in several orientations and scales. 3D DMD has several advantages over other methods: higher efficiency compared to existing methods; minimal computational costs when using an integrated image; low dimension; ease of implementation; does not require settings. The presented modification allows to increase productivity by 2-4%.

关键词：

Beyond background-aware correlation filters: Adaptive context modeling by hand-crafted and deep RGB features for visual tracking

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Marvasti-Zadeh, Seyed Mojtaba Ghanei-Yakhdan, Hossein Kasaei, Shohreh Department of Electrical Engineering Yazd University Yazd Iran Sharif University of Technology Tehran Iran Vision and Learning Lab. University of Alberta Edmonton Canada Department of Electrical Engineering Yazd University Yazd Iran Department of Computer Engineering Sharif University of Technology Tehran Iran

In recent years, the background-aware correlation filters have achieved a lot of research interest in the visual target tracking. However, these methods cannot suitably model the target appearance due to the exploitation of hand-crafted features. On the other hand, the recent deep learning-based visual tracking methods have provided a competitive performance along with extensive computations. In this paper, an adaptive background-aware correlation filter-based tracker is proposed that effectively models the target appearance by using either the histogram of oriented gradients (HOG) or convolutional neural network (CNN) feature maps. The proposed method exploits the fast 2D non-maximum suppression (NMS) algorithm and the semantic information comparison to detect challenging situations. When the HOG-based response map is not reliable, or the context region has a low semantic similarity with prior regions, the proposed method constructs the CNN context model to improve the target region estimation. Furthermore, the rejection option allows the proposed method to update the CNN context model only on valid regions. Comprehensive experimental results demonstrate that the proposed adaptive method clearly outperforms the accuracy and robustness of visual target tracking compared to the state-of-the-art methods on the OTB-50, OTB-100, TC-128, UAV-123, and VOT-2015 datasets. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep neural networks

TTPP: Temporal transformer with progressive prediction for efficient action anticipation

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Wang, Wen Peng, Xiaojiang Su, Yanzhou Qiao, Yu Cheng, Jian School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu Sichuan611731 China ShenZhen Key Lab of Computer Vision and Pattern Recognition SIAT-SenseTime Joint Lab. Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences SIAT Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society

Video action anticipation aims to predict future action categories from observed frames. Current state-of-the-art approaches mainly resort to recurrent neural networks to encode history information into hidden states, and predict future actions from the hidden representations. It is well known that the recurrent pipeline is inefficient in capturing long-term information which may limit its performance in predication task. To address this problem, this paper proposes a simple yet efficient Temporal Transformer with Progressive Prediction (TTPP) framework, which repurposes a Transformer-style architecture to aggregate observed features, and then leverages a light-weight network to progressively predict future features and actions. Specifically, predicted features along with predicted probabilities are accumulated into the inputs of subsequent prediction. We evaluate our approach on three action datasets, namely TVSeries, THUMOS-14, and TV-Human-Interaction. Additionally we also conduct a comprehensive study for several popular aggregation and prediction strategies. Extensive results show that TTPP not only outperforms the state-of-the-art methods but also more efficient. Copyright © 2020, The Authors. All rights reserved.

关键词： Forecasting

The quaternion-based anisotropic gradient for the color images 17

学校读者我要写书评

暂无评论

The quaternion-based anisotropic gradient for the color imag...

17th Image Processing: Algorithms and Systems Conference, IPAS 2019

作者： Voronin, V. Zelensky, A. Agaian, S. Don State Technical University Lab. «Mathematical Methods of Image Processing and Computer Vision Intelligent Systems» Gagarina 1 Rostov on Don Russia Moscow State University of Technology"STANKIN" Moscow Russia CUNY College of Staten Island Dept. of Computer Science New York United States

Image gradient, as a preprocessing step is an essential tool in image processing in many research areas such as edge detection, segmentation, smoothing, inpainting, etc. In the present paper, we develop a new gradient by integrated the quaternion framework with LPA-ICI (local polynomial approximation - the intersection of confidence intervals) based on anisotropic gradient concepts for the color image processing applications. computer simulations on the Berkeley Segmentation Dataset show that the new quaternion anisotropic gradient exhibits fewer color artifacts compared to state-of-the-art techniques including well-knowledge gradient operators. © 2019, Society for Imaging Science and technology.

关键词： Polynomial approximation

Feedback alfa-rooting algorithm for medical image enhancement 17

学校读者我要写书评

暂无评论

Feedback alfa-rooting algorithm for medical image enhancemen...

17th Image Processing: Algorithms and Systems Conference, IPAS 2019

作者： Voronin, V. Zelensky, A. Agaian, S. Don State Technical University Lab. «Mathematical Methods of Image Processing and Computer Vision Intelligent Systems» Gagarina 1 Rostov on Don Russia Moscow State University of Technology "STANKIN" Moscow Russia CUNY/ College of Staten Island Dept. of Computer Science New York United States

This paper presents a new combined local and global transform domain-based feedback image enhancement algorithm for medical diagnosis, treatment, and clinical research. The basic idea in using local alfa-rooting method is to apply it to different disjoint blocks with different sizes. The block size and alfa-rooting parameters driven through optimization using the Agaian's cost function (image enhancement non-reference quality measure). The presented new approach allows enhancing MRI and CT images with uneven lighting and brightness gradient by preserving the local image features/details. Extensive computer simulations (CS) on real medical images are offered to gage the presented method. CS shows that our method improves the contrast and enhances the details of the medical images effectively compared with the current state-of-art methods. © 2019, Society for Imaging Science and technology.

关键词： Clinical research

Generating 3D TOF-MRA volumes and segmentation lab.ls using generative adversarial networks

学校读者我要写书评

暂无评论

Medical Image Analysis 2022年 78卷 102396-102396页

作者： Subramaniam, Pooja Kossen, Tabea Ritter, Kerstin Hennemuth, Anja Hildebrand, Kristian Hilbert, Adam Sobesky, Jan Livne, Michelle Galinovic, Ivana Khalil, Ahmed A. Fiebach, Jochen B. Frey, Dietmar Madai, Vince I. CLAIM - Charité Lab for AI in Medicine Charité Universitätsmedizin Berlin Germany Department of Computer Engineering and Microelectronics Computer Vision & Remote Sensing Technical University Berlin Berlin Germany Berlin Germany Bernstein Center for Computational Neuroscience Berlin Germany Institute for Imaging Science and Computational Modelling in Cardiovascular Medicine Charité Universitätsmedizin Berlin Berlin Germany Fraunhofer MEVIS Max-von-Laue-Str. 2 Bremen Germany Department VI Computer Science and Media Beuth University of Applied Sciences Berlin Germany Johanna-Etienne-Hospital Neuss Germany Centre for Stroke Research Berlin Charité Universitätsmedizin Berlin Berlin Germany Department of Neurology Max Planck Institute for Human Cognitive and Brain Sciences Leipzig Germany Mind Brain Body Institute Berlin School of Mind and Brain Humboldt University Berlin Berlin Germany Berlin Institute of Health Berlin Germany School of Computing and Digital Technology Faculty of Computing Engineering and the Built Environment Birmingham City University Birmingham United Kingdom QUEST-Center for Transforming Biomedical Research Berlin Institute of Health Charité Universitätsmedizin Berlin Charitéplatz 1 Berlin10117 Germany

Deep learning requires large lab.led datasets that are difficult to gather in medical imaging due to data privacy issues and time-consuming manual lab.ling. Generative Adversarial Networks (GANs) can alleviate these challenges enabling synthesis of shareable data. While 2D GANs have been used to generate 2D images with their corresponding lab.ls, they cannot capture the volumetric information of 3D medical imaging. 3D GANs are more suitable for this and have been used to generate 3D volumes but not their corresponding lab.ls. One reason might be that synthesizing 3D volumes is challenging owing to computational limitations. In this work, we present 3D GANs for the generation of 3D medical image volumes with corresponding lab.ls applying mixed precision to alleviate computational constraints. We generated 3D Time-of-Flight Magnetic Resonance Angiography (TOF-MRA) patches with their corresponding brain blood vessel segmentation lab.ls. We used four variants of 3D Wasserstein GAN (WGAN) with: 1) gradient penalty (GP), 2) GP with spectral normalization (SN), 3) SN with mixed precision (SN-MP), and 4) SN-MP with double filters per layer (c-SN-MP). The generated patches were quantitatively evaluated using the Fréchet Inception Distance (FID) and Precision and Recall of Distributions (PRD). Further, 3D U-Nets were trained with patch-lab.l pairs from different WGAN models and their performance was compared to the performance of a benchmark U-Net trained on real data. The segmentation performance of all U-Net models was assessed using Dice Similarity Coefficient (DSC) and balanced Average Hausdorff Distance (bAVD) for a) all vessels, and b) intracranial vessels only. Our results show that patches generated with WGAN models using mixed precision (SN-MP and c-SN-MP) yielded the lowest FID scores and the best PRD curves. Among the 3D U-Nets trained with synthetic patch-lab.l pairs, c-SN-MP pairs achieved the highest DSC (0.841) and lowest bAVD (0.508) compared to the benchmark

关键词： Generative adversarial networks