检索结果-内蒙古大学图书馆

International Conference on Computer Vision (ICCV)

作者： Weilai Xiang Hongyu Yang Di Huang Yunhong Wang State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China Institute of Artificial Intelligence Beihang University Beijing China

Inspired by recent advances in diffusion models, which are reminiscent of denoising autoencoders, we investigate whether they can acquire discriminative representations for classification via generative pre-training. This paper shows that the networks in diffusion models, namely denoising diffusion autoencoders (DDAE), are unified self-supervised learners: by pre-training on unconditional image generation, DDAE has already learned strongly linear-separable representations within its intermediate layers without auxiliary encoders, thus making diffusion pre-training emerge as a general approach for generative-and-discriminative dual learning. To validate this, we conduct linear probe and finetuning evaluations. Our diffusion-based approach achieves 95.9% and 50.0% linear evaluation accuracies on CIFAR-10 and Tiny-ImageNet, respectively, and is comparable to contrastive learning and masked autoencoders for the first time. Transfer learning from ImageNet also confirms the suitability of DDAE for Vision Transformers, suggesting the potential to scale DDAEs as unified foundation models. Code is available at ***/FutureXiang/ddae.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Brightness function across varying adapting luminance in head-mounted displays

引用

Journal of the Optical Society of America A 2025年第7期42卷 927-937页

作者： Zan Jin Shining Ma Yue Liu Yongtian Wang Weitao Song State Key Laboratory of Chips and Systems for Advanced Light Field Display School of Optics and Photonics Beijing Institute of Technology Beijing 100081 China Beijing Engineering Research Center of Mixed Reality and Advanced Display School of Optics and Photonics Beijing Institute of Technology Beijing 100081 China

Brightness functions describe the relationship between perceived brightness and stimulus luminance under various viewing conditions. Previous studies have highlighted the effects of adapting luminance and ambient conditions on perceived brightness. Brightness functions are essential for ensuring accurate tone reproduction across different media and viewing conditions. Most prior research on this topic has been conducted in real environments using either simple or complex stimuli. With the development of head-mounted displays (HMDs), the need for accurate tone reproduction in 3D virtual scenes has emerged, requiring accurate brightness functions for different adapting luminance levels. However, brightness perception in HMDs has been scarcely investigated, and it remains unclear whether traditional brightness functions are applicable to HMDs. To address this gap, this study employed a magnitude estimation method to collect brightness ratings for stimuli in complex images under varying adapting luminance levels. The results revealed a systematic deviation between the observed brightness ratings and the model predictions, with stimuli appearing brighter in HMDs, particularly under low adapting luminance levels. This finding suggests that viewing conditions in HMDs—likely influenced by their limited field of view—affect brightness perception compared to real-world scenes.

关键词： Augmented reality Image enhancement Image metrics Image quality Photometry virtual reality

来源：评论

学校读者我要写书评

暂无评论

Robust Adaptive Filter for Time-Varying Parameters Estimation in Integrated Navigation of Fixed-Wing Aerial Robot*

Robust Adaptive Filter for Time-Varying Parameters Estimatio...

引用

IEEE International Conference on robotics and Biomimetics

作者： Zhaoyu Zhang Haibin Duan School of Automation Science and Electrical Engineering Beihang University Beijing PR China State Key Laboratory of Virtual Reality Technology and Systems School of Automation Science and Electrical Engineering Beihang University Beijing China Department of Mathematics and Theories Virtual Reality Fundamental Research Laboratory Peng Cheng Laboratory Shenzhen China

This paper focuses on the autonomous navigation problem of the fixed-wing aerial robot, of which the inertial measurement unit (IMU) has time-varying error characteristic parameters. A robust variational Bayesian adaptive Kalman filter (RVBAKF) is proposed to tackle the noise outliers in integrated navigation brought by sensor uncertainties. The proposed method formulates the joint probability distribution of the system state vector and the measurement noise covariance matrix. A fading factor matrix with strong tracking quality is introduced to enhance the prediction on the prior distribution of the process noise covariance. Then a weighted sliding window mechanism has been constructed to obtain the posterior distribution of the measurement noise outliers. Therefore, the proposed approach is impressive in approximating both the process and measurement noise. The RVBAKF algorithm is implemented in an integrated navigation system which is composed of the strapdown inertial navigation system (SINS) and the global navigation satellite system (GNSS). The integration framework based on RVBAKF is conducted on two typical verification scenarios, which is proven to be exceptional in coping with the time-varying process and measurement noise by comparing the average root mean square error in misalignment angle, velocity and position with the strong tracking filter and the variational Bayesian filter.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Towards Depth-continuous Scene Representation with Displacement Field for Robust Light Field Depth Estimation

引用

IEEE Transactions on Multimedia 2025年 27卷 3637-3649页

作者： Chen, Rongshan Sheng, Hao Yang, Da Cong, Ruixuan Cui, Zhenglong Wang, Sizhe Wang, Tun Zhao, Mingyuan Beihang University State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beijing100191 China Beihang University Key Laboratory of Data Science and Intelligent Computing International Innovation Institute 166 Shuanghongqiao Street Pingyao Town Yuhang District Hangzhou31 1115 China Macao Polytechnic University Faculty of Applied Sciences Macao999078 China

The light field (LF) captures both the spatial and angular information of scenes, enabling accurate depth estimation. However, previous deep learning methods typically model surface depth only while ignoring the continuous nature of depth in 3D scenes. In this paper, we use the displacement field (DF) to describe this continuous property, and propose a novel depth-continuous scene representation for robust LF depth estimation. Experiments demonstrate that our representation enables the network to generate highly detailed depth maps with fewer parameters and faster speeds. Specifically, inspired by the signed distance field in 3D object descriptions, we aim to exploit the intrinsic depth-continuous property of 3D scenes via DF and define a novel depth-continuous scene representation. Then, we introduce a simple, yet general, learning framework for depth-continuous scene embedding, and the proposed network, DepthDF, achieves state-of-the-art performance on both synthetic and real-world LF datasets, ranking 1st on the HCI 4D light field benchmark. Furthermore, previous LF depth estimation methods can also be seamlessly integrated into this framework. Finally, we extend this framework beyond LF depth estimation to various tasks, including multiview stereo depth inference, LF superresolution, and LF salient object detection. Our experiments demonstrate an improved performance when continuous scene representation is applied, suggesting that our framework can potentially provide insights for more fields. © 1999-2012 IEEE.

关键词： Stereo image processing

来源：评论

学校读者我要写书评

暂无评论

AR-based Dynamic Sandbox for Hydraulic Earth-rock Dam Break

AR-based Dynamic Sandbox for Hydraulic Earth-rock Dam Break

引用

virtual reality and 3D User Interfaces Abstracts and Workshops (VRW), IEEE Conference on

作者： Huiwen Liu Yuan Wang Dawei Zhang Yang Gao China Institute of Water Resources and Hydropower Research Beijing China POWERCHINA Beijing Engineering Corporation Limited Beijing China School of software Beihang University Beijing China State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China

ISBN: (数字)9798350374490

ISBN: (纸本)9798350374506

VRiAR-based hydraulic scenes holds significant value for geological disaster display. However, the river dynamics and dam break models prevalent in computational fluid dynamics (CFD) and hydraulics rely largely on pure numerical simulations, lacking intuitive visual representation. In contrast, current AR-based physical simulation methods often fail to incorporate hydraulic parameters adequately, falling short of the accuracy needed for real hydraulic models. To bridge this gap, this paper proposes integrating a precise numerical river dynamics model into a visual fluid-solid interaction simulation framework, which achieves a balance between a realistic visual simulation of dam breaks and the numerical accuracy aligned with river dynamics models, while maintaining high computational efficiency.

关键词： Visualization Solid modeling Dams Computational modeling Computational fluid dynamics Geology Hydraulic systems

来源：评论

学校读者我要写书评

暂无评论

Few-Shot Open-Set Object Detection via Conditional Evidence Decoupling

arXiv

引用

arXiv 2024年

作者： Wu, Zhaowei Su, Binyi Zhang, Hua Zhou, Zhong The State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science and Engineering Beihang University Beijing100191 China The School of Artificial Intelligence and Data Science Hebei University of Technology Tianjin300401 China The State Key Laboratory of Information Security Institute of Information Engineering Chinese Academy of Sciences Beijing100093 China Zhongguancun Laboratory Beijing100190 China

Few-shot Open-set Object Detection (FOOD) poses a significant challenge in real-world scenarios. It aims to train an open-set detector under the condition of scarce training samples, which can detect known objects while rejecting unknowns. Under this challenging scenario, the decision boundaries of unknowns are difficult to learn and often ambiguous. To mitigate this issue, we develop a two-stage open-set object detection framework with prompt learning, which delves into conditional evidence decoupling for the unknown rejection. Specifically, we propose an Attribution-Gradient-based Pseudo-unknown Mining (AGPM) method to select region proposals with high uncertainty, which leverages the discrepancy in attribution gradients between known and unknown classes, alleviating the inadequate unknown distribution coverage of training data. Subsequently, we decouple known and unknown properties in pseudo-unknown samples to learn distinct knowledge with proposed Conditional Evidence Decoupling (CED), which enhances separability between knowns and unknowns. Additionally, we adjust the output probability distribution through Abnormal Distribution Calibration (ADC), which serves as a regularization term to establish robust decision boundaries for the unknown rejection. Our method has achieved superior performance over previous state-of-the-art approaches, improving the mean recall of unknown class by 7.24% across all shots in VOC10-5-5 dataset settings and 1.38% in VOC-COCO dataset settings. Copyright © 2024, The Authors. All rights reserved.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Link-identified Routing Architecture in Space

arXiv

引用

arXiv 2024年

作者： Zhang, Hefan Wang, Zhiyuan Zhang, Shan Meng, Qingkai Luo, Hongbin The School of Computer Science and Engineering Beihang University Beijing100191 China The State Key Laboratory of Virtual Reality Technology and Systems Beijing100191 China The State Key Laboratory of Software Development Environment Beijing100191 China The Institute of Artificial Intelligence Beihang University Beijing100191 China The School of Cyber Science and Technology Beihang University Beijing100191 China

Low earth orbit (LEO) satellite networks have the potential to provide low-latency communication with global coverage. To unleash this potential, it is crucial to achieve efficient packet delivery. In this paper, we propose a Link-identified Routing (LiR) architecture for LEO satellite networks. The LiR architecture leverages the deterministic neighbor relation of LEO constellations, and identifies each inter-satellite link (ISL). Moreover, LiR architecture adopts source-route-style forwarding based on in-packet bloom filter (BF). Each satellite could efficiently encode multiple ISL identifiers via an in-packet BF to specify the end-to-end path for the packets. Due to false positives caused by BF, the more ISLs are encoded at a time, the more redundant forwarding cases emerge. Based on the topology characteristics, we derive the expected forwarding overhead in a closed-form and propose the optimal encoding policy. To accommodate link-state changes in LEO satellite networks, we propose the on-demand rerouting scheme and the on-demand detouring scheme to address the intermittent ISLs. We also elaborate how to take advantage of LiR architecture to achieve seamless handover for ground-satellite links (GSLs). Finally, we conduct extensive numerical experiments and packet-level simulations to verify our analytical results and evaluate the performance of the LiR architecture. © 2024, CC BY.

关键词： Satellite links

来源：评论

学校读者我要写书评

暂无评论

Out-of-Distribution Detection in Long-Tailed Recognition with Calibrated Outlier Class Learning

arXiv

引用

arXiv 2023年

作者： Miao, Wenjun Pang, Guansong Li, Tianqi Bai, Xiao Zheng, Jin School of Computer Science and Engineering Beihang University China School of Computing and Information Systems Singapore Management University Singapore State Key Laboratory of Software Development Environment Jiangxi Research Institute Beihang University China State Key Laboratory of Virtual Reality Technology and Systems Beihang University China

Existing out-of-distribution (OOD) methods have shown great success on balanced datasets but become ineffective in long-tailed recognition (LTR) scenarios where 1) OOD samples are often wrongly classified into head classes and/or 2) tail-class samples are treated as OOD samples. To address these issues, current studies fit a prior distribution of auxiliary/pseudo OOD data to the long-tailed in-distribution (ID) data. However, it is difficult to obtain such an accurate prior distribution given the unknowingness of real OOD samples and heavy class imbalance in LTR. A straightforward solution to avoid the requirement of this prior is to learn an outlier class to encapsulate the OOD samples. The main challenge is then to tackle the aforementioned confusion between OOD samples and head/tail-class samples when learning the outlier class. To this end, we introduce a novel calibrated outlier class learning (COCL) approach, in which 1) a debiased large margin learning method is introduced in the outlier class learning to distinguish OOD samples from both head and tail classes in the representation space and 2) an outlier-class-aware logit calibration method is defined to enhance the long-tailed classification confidence. Extensive empirical results on three popular benchmarks CIFAR10-LT, CIFAR100-LT, and ImageNet-LT demonstrate that COCL substantially outperforms state-of-the-art OOD detection methods in LTR while being able to improve the classification accuracy on ID data. Code is available at https://***/mala-lab/COCL. Copyright © 2023, The Authors. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

A New Dynamic Routing Network for Monocular Depth Estimation 12

A New Dynamic Routing Network for Monocular Depth Estimation

引用

12th International Conference on CYBER technology in Automation, Control, and Intelligent systems, CYBER 2022

作者： Luo, Zhehao Luo, Sijin Liang, Guoyuan Wu, Xinyu Guangdong Provincial Key Lab of Robotics and Intelligent System Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Province Shenzhen China University of Chinese Academy of Sciences Beijing China Guangdong-Hong Kong-Macao Joint Laboratory of Human-Machine Intelligence-Synergy Systems Chinese Academy of Sciences Guangdong Province Shenzhen China Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Province Shenzhen China

ISBN: (数字)9781665472678

ISBN: (纸本)9781665472678

Depth estimation is an essential task for understanding the geometry of 3D scenes. Compared with multi-view-based methods, monocular depth estimation is more challenging for the requirement of integrating not only global information but also local cues from various parts. Recently, numerous Convolutional Neural Networks (CNNs) based approaches relying on pre-defined static architectures have been reported. Although many of them have achieved remarkable improvements in accuracy, there remains problem with the generality of the static model when it comes to diverse scenes. In this paper, a novel deep network with dynamic routing architecture is proposed with a dynamically adjustable structure with respect to different inputs, therefore leading to not only improvements on both accuracy and generality but also decrease of the model parameters. Extensive experiments of monocular depth estimation have been conducted on two well-known challenging datasets (i.e., NYUD-v2 and Make3D) as well as the mixed dataset, and have verified the effectiveness of the proposed dynamic framework. © 2022 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Complex Object Illumination Transfer Through Semantic and Material Parsing and Composition 3rd

Complex Object Illumination Transfer Through Semantic and Ma...

引用

3rd EAI International Conference on Robotic Sensor Networks, ROSENET 2019

作者： Li, Xiaodong Han, Rui Ning, Ning Zhang, Xiaokun Jin, Xin Beijing Electronic Science and Technology Institute Beijing China CETC Big Data Research Institute Co. Ltd. Guizhou China State Key Laboratory of Virtual Reality Technology and Systems Beihang University Beijing China

ISBN: (纸本)9783030460310

Complex object illumination transfer presents a particular challenge in computer vision. We took up the challenge and proposed a novel complex object illumination transfer method through semantic and material parsing and composition. Firstly, we parsed the subject complex object into several parts through semantic parsing. We could easier match similar common objects that own same material and similar geometry with subject object parts as example objects through material parsing. For each subject object part, we need two example objects, one is same illumination to the subject object part, and the other is example illumination. Then, we warped two example objects to each corresponding subject object part through patch match. The example illumination transfer to each subject object part through the local and global transfer model learned from two warped example objects. Lastly, we combined each relighting subject object part into relighting subject object through semantic and material composition. Our results reveal that the method could produce similar illumination as the relighting ground truth subject object. © 2021, Springer Nature Switzerland AG.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：