检索结果-内蒙古大学图书馆

Cross-Ray Neural Radiance Fields for Novel-view Synthesis from Unconstrained Image Collections

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yang, Yifan Zhang, Shuhai Huang, Zixiong Zhang, Yubing Tan, Mingkui South China University of Technology China Pazhou Lab China Guangzhou Shiyuan Electronics Co. Ltd China Key Laboratory of Big Data and Intelligent Robot Ministry of Education China

Neural Radiance Fields (NeRF) is a revolutionary approach for rendering scenes by sampling a single ray per pixel and it has demonstrated impressive capabilities in novel-view synthesis from static scene images. However, in practice, we usually need to recover NeRF from unconstrained image collections, which poses two challenges: 1) the images often have dynamic changes in appearance because of different capturing time and camera settings;2) the images may contain transient objects such as humans and cars, leading to occlusion and ghosting artifacts. Conventional approaches seek to address these challenges by locally utilizing a single ray to synthesize a color of a pixel. In contrast, humans typically perceive appearance and objects by globally utilizing information across multiple pixels. To mimic the perception process of humans, in this paper, we propose Cross-Ray NeRF (CR-NeRF) that leverages interactive information across multiple rays to synthesize occlusion-free novel views with the same appearances as the images. Specifically, to model varying appearances, we first propose to represent multiple rays with a novel cross-ray feature and then recover the appearance by fusing global statistics, i.e., feature covariance of the rays and the image appearance. Moreover, to avoid occlusion introduced by transient objects, we propose a transient objects handler and introduce a grid sampling strategy for masking out the transient objects. We theoretically find that leveraging correlation across multiple rays promotes capturing more global information. Moreover, extensive experimental results on large real-world datasets verify the effectiveness of CR-NeRF. Copyright © 2023, The Authors. All rights reserved.

关键词： Pixels

Towards Stable Test-Time Adaptation in Dynamic Wild World

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Niu, Shuaicheng Wu, Jiaxiang Zhang, Yifan Wen, Zhiquan Chen, Yaofo Zhao, Peilin Tan, Mingkui South China University of Technology China Tencent AI Lab United States National University of Singapore Singapore Key Laboratory of Big Data and Intelligent Robot Ministry of Education China Pazhou Laboratory China

Test-time adaptation (TTA) has shown to be effective at tackling distribution shifts between training and testing data by adapting a given model on test samples. However, the online model updating of TTA may be unstable and this is often a key obstacle preventing existing TTA methods from being deployed in the real world. Specifically, TTA may fail to improve or even harm the model performance when test data have: 1) mixed distribution shifts, 2) small batch sizes, and 3) online imbalanced label distribution shifts, which are quite common in practice. In this paper, we investigate the unstable reasons and find that the batch norm layer is a crucial factor hindering TTA stability. Conversely, TTA can perform more stably with batch-agnostic norm layers, i.e., group or layer norm. However, we observe that TTA with group and layer norms does not always succeed and still suffers many failure cases. By digging into the failure cases, we find that certain noisy test samples with large gradients may disturb the model adaption and result in collapsed trivial solutions, i.e., assigning the same class label for all samples. To address the above collapse issue, we propose a sharpness-aware and reliable entropy minimization method, called SAR, for further stabilizing TTA from two aspects: 1) remove partial noisy samples with large gradients, 2) encourage model weights to go to a flat minimum so that the model is robust to the remaining noisy samples. Promising results demonstrate that SAR performs more stably over prior methods and is computationally efficient under the above wild test scenarios. The source code is available at https://***/mr-eggplant/SAR. Copyright © 2023, The Authors. All rights reserved.

关键词： Entropy

PathReasoner: Modeling Reasoning Path with Equivalent Extension for Logical Question Answering

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Xu, Fangzhi Lin, Qika Zhao, Tianzhe Han, Jiawei Liu, Jun School of Computer Science and Technology Xi’an Jiaotong University China Ministry of Education Key Laboratory of Intelligent Networks and Network Security China Shaanxi Province Key Laboratory of Big Data Knowledge Engineering China

Logical reasoning task has attracted great interest since it was proposed. Faced with such a task, current competitive models, even large language models (e.g., ChatGPT and PaLM 2), still perform badly. Previous promising LMs struggle in logical consistency modeling and logical structure perception. To this end, we model the logical reasoning task by transforming each logical sample into reasoning paths and propose an architecture PathReasoner. It addresses the task from the views of both data and model. To expand the diversity of the logical samples, we propose an atom extension strategy supported by equivalent logical formulas, to form new reasoning paths. From the model perspective, we design a stack of transformer-style blocks. In particular, we propose a path-attention module to joint model in-atom and cross-atom relations with the high-order diffusion strategy. Experiments show that PathReasoner achieves competitive performances on two logical reasoning benchmarks and great generalization abilities. © 2024, CC BY.

关键词： Atoms

Score-based Generative Priors Guided Model-driven Network for MRI Reconstruction

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Qiao, Xiaoyu Li, Weisheng Xiao, Bin Huang, Yuping Yang, Lijian Chongqing Key Laboratory of Image Cognition Chongqing University of Posts and Telecommunications Chongqing China Key Laboratory of Cyberspace Big Data Intelligent Security Chongqing University of Posts and Telecommunications Ministry of Education Chongqing China

Score matching with Langevin dynamics (SMLD) method has been successfully applied to accelerated MRI. However, the hyperparameters in the sampling process require subtle tuning, otherwise the results can be severely corrupted by hallucination artifacts, especially with out-of-distribution test data. To address the limitations, we proposed a novel workflow where naive SMLD samples serve as additional priors to guide model-driven network training. First, we adopted a pretrained score network to generate samples as preliminary guidance images (PGI), obviating the need for network retraining, parameter tuning and in-distribution test data. Although PGIs are corrupted by hallucination artifacts, we believe they can provide extra information through effective denoising steps to facilitate reconstruction. Therefore, we designed a denoising module (DM) in the second step to coarsely eliminate artifacts and noises from PGIs. The features are extracted from a score-based information extractor (SIE) and a cross-domain information extractor (CIE), which directly map to the noise patterns. Third, we designed a model-driven network guided by denoised PGIs (DGIs) to further recover fine details. DGIs are densely connected with intermediate reconstructions in each cascade to enrich the information and are periodically updated to provide more accurate guidance. Our experiments on different datasets reveal that despite the low average quality of PGIs, the proposed workflow can effectively extract valuable information to guide the network training, even with severely reduced training data and sampling steps. Our method outperforms other cutting-edge techniques by effectively mitigating hallucination artifacts, yielding robust and high-quality reconstruction results. Copyright © 2024, The Authors. All rights reserved.

关键词： Compressed sensing

IF-Font: ideographic description sequence-following font generation 24

学校读者我要写书评

暂无评论

IF-Font: ideographic description sequence-following font gen...

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Xinping Chen Xiao Ke Wenzhong Guo Fujian Provincial Key Laboratory of Networking Computing and Intelligent Information Processing College of Computer and Data Science Fuzhou University Fuzhou China and Engineering Research Center of Big Data Intelligence Ministry of Education Fuzhou China

ISBN: (纸本)9798331314385

Few-shot font generation (FFG) aims to learn the target style from a limited number of reference glyphs and generate the remaining glyphs in the target font. Previous works focus on disentangling the content and style features of glyphs, combining the content features of the source glyph with the style features of the reference glyph to generate new glyphs. However, the disentanglement is challenging due to the complexity of glyphs, often resulting in glyphs that are influenced by the style of the source glyph and prone to artifacts. We propose IF-Font, a novel paradigm which incorporates Ideographic Description Sequence (IDS) instead of the source glyph to control the semantics of generated glyphs. To achieve this, we quantize the reference glyphs into tokens, and model the token distribution of target glyphs using corresponding IDS and reference tokens. The proposed method excels in synthesizing glyphs with neat and correct strokes, and enables the creation of new glyphs based on provided IDS. Extensive experiments demonstrate that our method greatly outperforms state-of-the-art methods in both one-shot and few-shot settings, particularly when the target styles differ significantly from the training font styles. The code is available at https://***/Stareven233/IF-Font.

关键词：

Advanced Detection and Signal Reception System Utilizing Rydberg Atomic Dynamics for Applications

学校读者我要写书评

暂无评论

Advanced Detection and Signal Reception System Utilizing Ryd...

European Conference on Antennas and Propagation, EuCAP

作者： Wenjie Ding Guoda Xie Yingsong Li Zhixiang Huang The Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University Hefei China Ministry of Education Center for Big Data and Population Health of IHM Anhui University Hefei China The Jianghuai Advance Technology Center Hefei China

ISBN: (数字)9788831299107

ISBN: (纸本)9798350366327

In this paper, we introduce a novel Monte Carlo wave function method (MCWFM) for simulating the transient response of Rydberg atoms under rapidly varying electric fields, aimed at high-precision electric field measurements and wireless communication reception. This approach leverages the electromagnetically induced transparency (EIT) effect and Autler- Townes (AT) splitting in Rydberg atoms, offering enhanced efficiency compared to traditional density matrix methods by focusing solely on wave function evolution, thereby reducing computational complexity. Additionally, we account for Doppler effects in AT splitting through MCWFM, providing faster results than conventional Runge-Kutta methods. This work lays a solid foundation for advancing Rydberg atom technology in the development of high-capacity, high-fidelity wireless communication systems.

关键词： Wireless communication Atomic measurements Transient response Rydberg atoms Receivers Wave functions Optical fiber communication Doppler effect Electric fields Computational complexity

Variable projection algorithms with sparse constraint for separable nonlinear models

学校读者我要写书评

暂无评论

Control Theory and Technology 2024年第1期22卷 135-146页

作者： Hui-Lang Xu Guang-Yong Chen Si-Qing Cheng Min Gan Jing Chen College of Computer and Data Science Fuzhou UniversityFuzhou350116FujianChina Fujian Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou UniversityFuzhou350116FujianChina Key Laboratory of Intelligent Metro of Universities in Fujian Fuzhou UniversityFuzhou350116FujianChina Engineering Research Center of Big Data Intelligence Ministry of EducationFuzhou UniversityFuzhou350116FujianChina School of Science Jiangnan UniversityWuxi214122JiangsuChina

Separable nonlinear models are widely used in various fields such as time series analysis, system modeling, and machine learning, due to their flexible structures and ability to capture nonlinear behavior of data. However, identifying the parameters of these models is challenging, especially when sparse models with better interpretability are desired by practitioners. Previous theoretical and practical studies have shown that variable projection (VP) is an efficient method for identifying separable nonlinear models, but these are based on \(L_2\) penalty of model parameters, which cannot be directly extended to deal with sparse constraint. Based on the exploration of the structural characteristics of separable models, this paper proposes gradient-based and trust-region-based variable projection algorithms, which mainly solve two key problems: how to eliminate linear parameters under sparse constraint;and how to deal with the coupling relationship between linear and nonlinear parameters in the model. Finally, numerical experiments on synthetic data and real time series data are conducted to verify the effectiveness of the proposed algorithms.

关键词： Variable projection(VP) Non-smooth constraint Separable nonlinear models

Densely packed object detection with transformer-based head and EM-merger

学校读者我要写书评

暂无评论

Densely packed object detection with transformer-based head ...

作者： Zhong, Xiaojing Zhang, Ni Hu, Hao Li, Li Cen, Junhua Wu, Qingyao School of Software Engineering South China University of Technology Guangzhou China Key Laboratory of Big Data and Intelligent Robot Ministry of Education Guangzhou China Pazhou Lab Guangzhou China Peng Cheng Laboratory Guangzhou China Zhongnan Building Materials Group Co. Ltd. Guangdong GW Guangzhou China Internet Technology Co. Ltd. Guangdong GW Guangzhou China

Due to the high density of objects and their varying sizes, detecting them accurately and without repetition in such scenarios is more challenging than traditional object detection methods. In this paper, we propose a YOLOv5-based object detection approach equipped with a Transformer-based Head and EM-Merger unit specifically designed for densely packed scenes. We incorporate the transformer architecture into the prediction heads to enable a self-attention mechanism that captures long-term dependencies between the densely packed objects. Additionally, we introduce an EM-Merger unit to resolve redundant object detections. Experimental results on the RebarDSC and SKU110K datasets demonstrate that our method significantly outperforms the baseline approach, achieving new state-of-the-art detection performance. © 2023, The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature.

关键词： Object detection

Using Depth-Enhanced Spatial Transformation for Student Gaze Target Estimation in Dual-View Classroom Images

学校读者我要写书评

暂无评论

Using Depth-Enhanced Spatial Transformation for Student Gaze...

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Miao, Haonan Zhao, Peizheng Sun, Yuqi Nan, Fang Zhang, Xiaolong Wu, Yaqiang Tian, Feng School of Computer Science and Technology Xi'an Jiaotong University Xi'an710049 China Ministry of Education Key Laboratory of Intelligent Networks and Network Security Xi'an Jiaotong University Xi'an710049 China School of Advanced Technology Xi'an Jiaotong-Liverpool University Suzhou215123 China Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi'an Jiaotong University Xi'an710049 China

ISBN: (纸本)9798350368741

Dual-view gaze target estimation in classroom environments has not been thoroughly explored. Existing methods lack consideration of depth information, primarily focusing on 2D image information and neglecting the latent 3D spatial context, which could lead to suboptimal transformation and cause the gaze cone to intersect with an incorrect object. This paper introduces a novel dual-view gaze target estimation method tailored for classroom settings, leveraging depth-enhanced spatial transformations. By formulating a depth-enhanced 2D space, our method uses depth-enhanced spatial transformation to accurately project students' gaze cones to the teacher-oriented image. Additionally, we collected a dataset named DVSGE, specifically for student gaze target estimation in dual-view classroom images. Experimental results demonstrate significant performance improvements of 9.8% in AUC and 19.9% in L2-Distance for our method, surpassing existing methods. © 2025 IEEE.

关键词： classroom depth-enhanced 2D space dual-view gaze target estimation spatial transformation