检索结果-内蒙古大学图书馆

LIQA: Lifelong blind image quality assessment

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Liu, Jianzhao Zhou, Wei Xu, Jiahua Li, Xin An, Shukun Chen, Zhibo CAS Key Laboratory of Technology in Geo-Spatial Information Processing and Application System University of Science and Technology of China Hefei230027 China

Existing blind image quality assessment (BIQA) methods are mostly designed in a disposable way and cannot evolve with unseen distortions adaptively, which greatly limits the deployment and application of BIQA models in real-world scenarios. To address this problem, we propose a novel Lifelong blind Image Quality Assessment (LIQA) approach, targeting to achieve the lifelong learning of BIQA. Without accessing to previous training data, our proposed LIQA can not only learn new distortions, but also mitigate the catastrophic forgetting of seen distortions. Specifically, we adopt the Split-and-Merge distillation strategy to train a single-head network that makes task-agnostic predictions. In the split stage, we first employ a distortion-specific generator to obtain the pseudo features of each seen distortion. Then, we use an auxiliary multi-head regression network to generate the predicted quality of each seen distortion. In the merge stage, we replay the pseudo features paired with pseudo labels to distill the knowledge of multiple heads, which can build the final regressed single head. Experimental results demonstrate that the proposed LIQA method can handle the continuous shifts of different distortion types and even datasets. More importantly, our LIQA model can achieve stable performance even if the task sequence is long. Copyright © 2021, The Authors. All rights reserved.

关键词： Distillation

Statistical analysis of typical target regions of multi-Aspect SAR based on parametric model 14

学校读者我要写书评

暂无评论

Statistical analysis of typical target regions of multi-Aspe...

14th European Conference on Synthetic Aperture Radar, EUSAR 2022

作者： Yue, Xiaoyang Teng, Fei Zhang, Zhiguo Lin, Yun Hong, Wen Key Laboratory of Spatial Information Processing and Application System Technology Chinese Academy of Sciences Beijing China Aerospace Information Research Institute Chinese Academy of Sciences Beijing China School of Electronic Electrical and Communication Engineering University of Chinese Academy of Sciences Beijing China School of Electronic Information Engineering North China University of Technology Beijing China

ISBN: (纸本)9783800758234

Multi-Aspect SAR can obtain rich backscatter information about the illuminated scene. For different kinds of targets in the scene, scattering information related to the geometric characteristics of the target is shown in the imaging scene at the same angle. For a certain reference target in the scene, it exhibits azimuth-based backscatter characteristics in the imaging scene under observation from different angles. In our article, a parametric statistical model is used to express characteristics of typical targets at different angles and characteristic differences between different targets. Two parameters-And-Are used to represent characteristic difference of typical targets. © 2022 Institute of Electrical and Electronics Engineers Inc.. All rights reserved.

关键词： Synthetic aperture radar

MASA: Motion-aware Masked Autoencoder with Semantic Alignment for Sign Language Recognition

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhao, Weichao Hu, Hezhen Zhou, Wengang Mao, Yunyao Wang, Min Li, Houqiang The CAS Key Laboratory of Technology in Geospatial Information Processing and Application System Department of Electronic Engineering and Information Science University of Science and Technology of China Hefei230027 China The Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230030 China

Sign language recognition (SLR) has long been plagued by insufficient model representation capabilities. Although current pre-training approaches have alleviated this dilemma to some extent and yielded promising performance by employing various pretext tasks on sign pose data, these methods still suffer from two primary limitations: i) Explicit motion information is usually disregarded in previous pretext tasks, leading to partial information loss and limited representation capability. ii) Previous methods focus on the local context of a sign pose sequence, without incorporating the guidance of the global meaning of lexical signs. To this end, we propose a Motion-Aware masked autoencoder with Semantic Alignment (MASA) that integrates rich motion cues and global semantic information in a self-supervised learning paradigm for SLR. Our framework contains two crucial components, i.e., a motion-aware masked autoencoder (MA) and a momentum semantic alignment module (SA). Specifically, in MA, we introduce an autoencoder architecture with a motion-aware masked strategy to reconstruct motion residuals of masked frames, thereby explicitly exploring dynamic motion cues among sign pose sequences. Moreover, in SA, we embed our framework with global semantic awareness by aligning the embeddings of different augmented samples from the input sequence in the shared latent space. In this way, our framework can simultaneously learn local motion cues and global semantic features for comprehensive sign language representation. Furthermore, we conduct extensive experiments to validate the effectiveness of our method, achieving new state-of-the-art performance on four public benchmarks. The source code are publicly available at https://***/sakura/MASA. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

Target Amplitude Characteristic Analysis With Neighborhood information Using C-band Circular SAR Data

学校读者我要写书评

暂无评论

Target Amplitude Characteristic Analysis With Neighborhood I...

2021 CIE International Conference on Radar, Radar 2021

作者： Yue, Xiaoyang Teng, Fei Lin, Yun Hong, Wen Areospace Information Research Institute Chinese Academy of Sciences Key Laboratory of Spatial Information Processing and Application System Technology Beijing China School of Electronic Information Engineering North China University of Technology Beijing China

ISBN: (纸本)9781665498142

The circular synthetic aperture radar (SAR) can observe the experimental scene from all angles. The backscatter intensity of the target in the scene can be obtained. Different targets in the imaging scene show different amplitude characteristics. For example, significant targets such as buildings and oil tanks show strong response characteristics at certain angles, and secondly, weak response characteristics for targets is shown such as grass and ground. In this article, the omnidirectional scattering intensity data of the target obtained from the radar is used to analyze the amplitude characteristics, and the model is established for different targets to obtain the final result. The fuzzy C-means (FCM) model combined with neighborhood information is used to analyze the target amplitude characteristics. The amplitude characteristics of the target are divided into two categories: strong response and weak response for analysis. The initial two types of cluster centers are set for iteration and finally the target amplitude characteristic reference values at different angles are obtained. C-band circular SAR data is used to validate our method. As a result, the amplitude characteristics of the whole scene using membership degree can be described. © 2021 IEEE.

关键词： amplitude circular FCM neighborhood target

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wu, Hui Wang, Min Zhou, Wengang Li, Houqiang The CAS Key Laboratory of Technology in Geospatial Information Processing and Application System Department of Electronic Engineering and Information Science University of Science and Technology of China Hefei230027 China The Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230030 China

Asymmetric image retrieval is a task that seeks to balance retrieval accuracy and efficiency by leveraging lightweight and large models for the query and gallery sides, respectively. The key to asymmetric image retrieval is realizing feature compatibility between different models. Despite the great progress, most existing approaches either rely on classifiers inherited from gallery models or simply impose constraints at the instance level, ignoring the structure of embedding space. In this work, we propose a simple yet effective structure similarity preserving method to achieve feature compatibility between query and gallery models. Specifically, we first train a product quantizer offline with the image features embedded by the gallery model. The centroid vectors in the quantizer serve as anchor points in the embedding space of the gallery model to characterize its structure. During the training of the query model, anchor points are shared by the query and gallery models. The relationships between image features and centroid vectors are considered as structure similarities and constrained to be consistent. Moreover, our approach makes no assumption about the existence of any labeled training data and thus can be extended to an unlimited amount of data. Comprehensive experiments on large-scale landmark retrieval demonstrate the effectiveness of our approach. Our code is released at: https://***/MCC-WH/SSP. © 2024, CC BY.

关键词： Embeddings

DouZero+: Improving DouDizhu AI by Opponent Modeling and Coach-guided Learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhao, Youpeng Zhao, Jian Hu, Xunhan Zhou, Wengang Li, Houqiang CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System Department of Electronic Engineering and Information Science University of Science and Technology of China Hefei230027 China Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei China

Recent years have witnessed the great breakthrough of deep reinforcement learning (DRL) in various perfect and imperfect information games. Among these games, DouDizhu, a popular card game in China, is very challenging due to the imperfect information, large state space, elements of collaboration and a massive number of possible moves from turn to turn. Recently, a DouDizhu AI system called DouZero has been proposed. Trained using traditional Monte Carlo method with deep neural networks and self-play procedure without the abstraction of human prior knowledge, DouZero has outperformed all the existing DouDizhu AI programs. In this work, we propose to enhance DouZero by introducing opponent modeling into DouZero. Besides, we propose a novel coach network to further boost the performance of DouZero and accelerate its training process. With the integration of the above two techniques into DouZero, our DouDizhu AI system achieves better performance and ranks top in the Botzone leaderboard among more than 400 AI agents, including DouZero. Copyright © 2022, The Authors. All rights reserved.

关键词： Reinforcement learning

TomoSAR-ALISTA: Efficient TomoSAR Imaging via Deep Unfolded Network

学校读者我要写书评

暂无评论

TomoSAR-ALISTA: Efficient TomoSAR Imaging via Deep Unfolded ...

2022 International Conference on Radar systems, RADAR 2022

作者： Wang, Muhan Zhang, Zhe Wang, Yue Gao, Silin Qiu, Xiaolan Key Laboratory of Technology in Geo-spatial Information Processing and Application System Chinese Academy of Sciences Beijing100190 China Key Laboratory of Intelligent Aerospace Big Data Application Technology Suzhou215123 China Suzhou Aerospace Information Research Institute Suzhou215123 China School of Electronic Electrical and Communication Engineering University of Chinese Academy of Sciences Beijing100049 China Aerospace Information Research Institute Chinese Academy of Sciences Beijing100094 China Electrical and Computer Engineering Department George Mason University FairfaxVA22030 United States

ISBN: (纸本)9781839537776

Synthetic aperture radar (SAR) tomography (TomoSAR) has attracted remarkable interest for its ability in achieving three-dimensional reconstruction along the elevation direction from multiple observations. In recent years, compressed sensing (CS) technique has been introduced into TomoSAR considering for its super-resolution ability with limited samples. Whereas, the CS-based methods suffer from several drawbacks, including weak noise resistance, high computational complexity and complex parameter fine-tuning. Among the different CS algorithms, iterative soft-thresholding algorithm (ISTA) is widely used as a robust reconstruction approach, however, the parameters in the ISTA are manually chosen, which usually requires a time-consuming fine-tuning process to achieve the best performance. Aiming at efficient TomoSAR imaging, a novel sparse unfolding network named analytic learned ISTA (ALISTA) is proposed towards the TomoSAR imaging problem in this paper, and the key parameters of ISTA are learned from training data via deep learning to avoid complex parameter fine-tuning and significantly relieves the training burden. In addition, experiments verify that it is feasible to use traditional CS algorithms as training labels, which provides a tangible supervised training method to achieve better 3D reconstruction performance even in the absence of labeled data in real applications. © 2022 IET Conference Proceedings. All rights reserved.

关键词： Compressed sensing

AF2S: An Anchor-Free Two-Stage Tracker Based on a Strong SiamFC Baseline 16th

学校读者我要写书评

暂无评论

AF2S: An Anchor-Free Two-Stage Tracker Based on a Strong Sia...

Workshops held at the 16th European Conference on Computer Vision, ECCV 2020

作者： He, Anfeng Wang, Guangting Luo, Chong Tian, Xinmei Zeng, Wenjun CAS Key Laboratory of Technology in Geo-Spatial Information Processing and Application System University of Science and Technology of China Hefei Anhui China Microsoft Research Asia Beijing China

ISBN: (纸本)9783030682378

Siamese network based trackers have become a mainstream in visual object tracking. Recently, several high-performance multi-stage trackers have been proposed and some of them adopt SiamRPN for the first-stage region proposal. We argue that an anchor-based region proposal network is not necessary for the tracking task, as a tracker has a strong prior about the location and size of the target. In this paper, we propose a two-stage visual tracker which uses SiamFC for region proposal. SiamFC defines a bounding box by its center, which is a typical anchor-free (AF) network, so we dub our tracker AF2S. As the model size of SiamFC is only about 1/10 that of SiamRPN, AF2S results in a significantly lighter model than its SiamRPN-based counterparts. In the design of AF2S, we first build a strong AlexNet-based SiamFC baseline which improves the AUC on OTB-100 from 0.582 to 0.665. Further, we propose a position-sensitive convolutional layer which can be stacked after SiamFC backbone to increase the robustness of proposals without losing localization precision. Finally, a relation network is used for box refinement. Experimental results show that AF2S achieves the best performance on OTB-100 and VOT-18 among the state-of-the-art trackers which use AlexNet as backbone. On LaSOT-test, AF2S achieves an AUC of 0.480, which is among the first-tier performance even when trackers with more powerful backbone and much larger model size are considered. © 2020, Springer Nature Switzerland AG.

关键词： Tracking (position)

Simultaneous Double Q-learning with Conservative Advantage Learning for Actor-Critic Methods

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Li, Qing Zhou, Wengang Lu, Zhenbo Li, Houqiang The CAS Key Laboratory of Technology in Geo-Spatial Information Processing and Application System Department of Electronic Engineering and Information Science University of Science and Technology of China Hefei230027 China The Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230000 China

Actor-critic Reinforcement Learning (RL) algorithms have achieved impressive performance in continuous control tasks. However, they still suffer two nontrivial obstacles, i.e., low sample efficiency and overestimation bias. To this end, we propose Simultaneous Double Q-learning with Conservative Advantage Learning (SDQ-CAL). Our SDQ-CAL boosts the Double Q-learning for off-policy actor-critic RL based on a modification of the Bellman optimality operator with Advantage Learning. Specifically, SDQ-CAL improves sample efficiency by modifying the reward to facilitate the distinction from experience between the optimal actions and the others. Besides, it mitigates the overestimation issue by updating a pair of critics simultaneously upon double estimators. Extensive experiments reveal that our algorithm realizes less biased value estimation and achieves state-of-the-art performance in a range of continuous control benchmark tasks. We release the source code of our method at: https://***/LQNew/SDQ-CAL. Copyright © 2022, The Authors. All rights reserved.

关键词： Reinforcement learning