检索结果-南通市图书馆

SGformer:Boosting transformers for indoor lighting estimation from a single image

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computational Visual Media 2024年第4期10卷 671-686页

作者： Junhong Zhao Bing Xue Mengjie Zhang Centre for Data science and Artificial Intelligence&school of engineering and computer science Victoria University of WellingtonWellington 6012New Zealand

Estimating lighting from standard images can effectively circumvent the need for resourceintensive high-dynamic-range(HDR)lighting ***,this task is often ill-posed and challenging,particularly for indoor scenes,due to the intricacy and ambiguity inherent in various indoor illumination *** propose an innovative transformer-based method called SGformer for lighting estimation through modeling spherical Gaussian(SG)distributions—a compact yet expressive lighting *** from previous approaches,we explore underlying local and global dependencies in lighting features,which are crucial for reliable lighting ***,we investigate the structural relationships spanning various resolutions of SG distributions,ranging from sparse to dense,aiming to enhance structural consistency and curtail potential stochastic noise stemming from independent SG component *** harnessing the synergy of local–global lighting representation learning and incorporating consistency constraints from various SG resolutions,the proposed method yields more accurate lighting estimates,allowing for more realistic lighting effects in object relighting and *** code and model implementing our work can be found at https://***/junhong-jennifer-zhao/SGformer.

关键词： lighting estimation transformer spherical Gaussian(SG) augmented reality

Focusing on your subject:Deep subject-aware image composition recommendation networks

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computational Visual Media 2023年第1期9卷 87-107页

作者： Guo-Ye Yang Wen-Yang Zhou Yun Cai Song-Hai Zhang Fang-Lue Zhang BNRist Department of Computer Science and TechnologyTsinghua UniversityBeiing 100084China school of engineering and computer science Victoria University of WellingtonWellington 6012New Zealand

Photo composition is one of the most important factors in the aesthetics of *** a popular application,composition recommendation for a photo focusing on a specific subject has been ignored by recent deep-learning-based composition recommendation *** this paper,we propose a subject-aware image composition recommendation method,SAC-Net,which takes an RGB image and a binary subject window mask as input,and returns good compositions as crops containing the *** model first determines candidate scores for all possible coarse cropping *** crops with high candidate scores are selected and further refined by regressing their corner points to generate the output recommended cropping *** final scores of the refined crops are predicted by a final score regression *** existing methods that need to preset several cropping windows,our network is able to automatically regress cropping windows with arbitrary aspect ratios and *** propose novel stability losses for maximizing smoothness when changing cropping windows along with view *** results show that our method outperforms state-of-the-art methods not only on the subject-aware image composition recommendation task,but also for general purpose composition *** also have designed a multistage labeling scheme so that a large amount of ranked pairs can be produced *** use this scheme to propose the first subject-aware composition dataset SACD,which contains 2777 images,and more than 5 million composition ranked *** SACD dataset is publicly available at https://***/SACD/.

关键词： subject-aware image composition image cropping deep learning recommendation

A Survey on 360°Images and Videos in Mixed Reality:Algorithms and Applications

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of computer science & Technology 2023年第3期38卷 473-491页

作者：张方略赵军红张赟 Stefanie Zollmann school of engineering and computer science Victoria University of WellingtonWellington 6012New Zealand College of Media engineering Communication University of ZhejiangHangzhou 310018China Department of computer science University of OtagoDunedin 9054New Zealand

Mixed reality technologies provide real-time and immersive experiences,which bring tremendous opportunities in entertainment,education,and enriched experiences that are not directly accessible owing to safety or *** research in this field has been in the spotlight in the last few years as the metaverse went *** recently emerging omnidirectional video streams,i.e.,360°videos,provide an affordable way to capture and present dynamic real-world *** the last decade,fueled by the rapid development of artificial intelligence and computational photography technologies,the research interests in mixed reality systems using 360°videos with richer and more realistic experiences are dramatically increased to unlock the true potential of the *** this survey,we cover recent research aimed at addressing the above issues in the 360°image and video processing technologies and applications for mixed *** survey summarizes the contributions of the recent research and describes potential future research directions about 360°media in the field of mixed reality.

关键词： 360°image mixed reality 360°image processing virtual reality scene reconstruction virtual reality content manipulation

A survey on deep geometry learning:From a representation perspective

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computational Visual Media 2020年第2期6卷 113-133页

作者： Yun-Peng Xiao Yu-Kun Lai Fang-Lue Zhang Chunpeng Li Lin Gao Institute of Computing Technology Chinese Academy of SciencesBeijingChina school of computer science and Informatics Cardiff UniversityWalesUK school of engineering and computer science Victoria University of WellingtonWellingtonNew Zealand

Researchers have achieved great success in dealing with 2 D images using deep *** recent years,3 D computer vision and geometry deep learning have gained ever more *** advanced techniques for 3 D shapes have been proposed for different *** 2 D images,which can be uniformly represented by a regular grid of pixels,3 D shapes have various representations,such as depth images,multi-view images,voxels,point clouds,meshes,implicit surfaces,*** performance achieved in different applications largely depends on the representation used,and there is no unique representation that works well for all ***,in this survey,we review recent developments in deep learning for 3 D geometry from a representation perspective,summarizing the advantages and disadvantages of different representations for different *** also present existing datasets in these representations and further discuss future research directions.

关键词： 3D shape representation geometry learning neural networks computer graphics

Image-based clothes changing system

在线全文

学校读者我要写书评

暂无评论

Computational Visual Media 2017年第4期3卷 337-346页

作者： Zhao-Heng Zheng Hao-Tian Zhang Fang-Lue Zhang Tai-Jiang Mu computer science and engineering University of Michigan computer science Department Stanford University school of engineering and computer science Victoria University of Wellington TNList Tsinghua University

Current image-editing tools do not match up to the demands of personalized image manipulation,one application of which is changing clothes in usercaptured images. Previous work can change single color clothes using parametric human warping *** this paper, we propose an image-based clothes changing system, exploiting body factor extraction and content-aware image warping. Image segmentation and mask generation are first applied to the user ***, we determine joint positions via a neural network. Then, body shape matching is performed and the shape of the model is warped to the user's shape. Finally, head swapping is performed to produce realistic virtual results. We also provide a supervision and labeling tool for refinement and further assistance when creating a dataset.

关键词： clothing try-on image warping human segmentation

JMNet: A joint matting network for automatic human matting

在线全文

学校读者我要写书评

暂无评论

Computational Visual Media 2020年第2期6卷 215-224页

作者： Xian Wu Xiao-Nan Fang Tao Chen Fang-Lue Zhang Department of computer science and Technology Tsinghua UniversityBeijing100084China AI Center at Visual China Group BurlingameCA94010USA school of engineering and computer science Victoria University of WellingtonWellingtonNew Zealand

We propose a novel end-to-end deep learning framework, the Joint Matting Network(JMNet), to automatically generate alpha mattes for human *** utilize the intrinsic structures of the human body as seen in images by introducing a pose estimation module,which can provide both global structural guidance and a local attention focus for the matting task. Our network model includes a pose network, a trimap network, a matting network, and a shared encoder to extract features for the above three networks. We also append a trimap refinement module and utilize gradient loss to provide a sharper alpha matte. Extensive experiments have shown that our method outperforms state-of-theart human matting techniques;the shared encoder leads to better performance and lower memory *** model can process real images downloaded from the Internet for use in composition applications.

关键词： alpha matting human images deep learning pose estimation

Robust tracking-by-detection using a selection and completion mechanism

在线全文

学校读者我要写书评

暂无评论

Computational Visual Media 2017年第3期3卷 285-294页

作者： Ruochen Fan Fang-Lue Zhang Min Zhang Ralph R.Martin Tsinghua university school of engineering and computer science Victoria University of Wellington Center of Mathematical sciences and Applications Harvard university school of computer science and Informatics Cardiff university

It is challenging to track a target continuously in videos with long-term occlusion,or objects which leave then re-enter a *** tracking algorithms combined with onlinetrained object detectors perform unreliably in complex conditions, and can only provide discontinuous trajectories with jumps in position when the object is occluded. This paper proposes a novel framework of tracking-by-detection using selection and completion to solve the abovementioned problems. It has two components, tracking and trajectory completion. An offline-trained object detector can localize objects in the same category as the object being tracked. The object detector is based on a highly accurate deep learning model. The object selector determines which object should be used to re-initialize a traditional tracker. As the object selector is trained online,it allows the framework to be adaptable. During completion, a predictive non-linear autoregressive neural network completes any discontinuous *** tracking component is an online real-time algorithm, and the completion part is an after-theevent mechanism. Quantitative experiments show a significant improvement in robustness over prior stateof-the-art methods.

关键词： object tracking detection proposal selection trajectory completion

RecStitchNet:Learning to stitch images with rectangular boundaries

在线全文

学校读者我要写书评

暂无评论

Computational Visual Media 2024年第4期10卷 687-703页

作者： Yun Zhang Yu-Kun Lai Lang Nie Fang-Lue Zhang Lin Xu Key Lab of Film and TV Media Technology of Zhejiang Province College of Media EngineeringCommunication University of ZhejiangHangzhou 310018China school of computer science and Informatics Cardiff UniversityCardiff CF244AGUK Institute of Information science Beijing Jiaotong UniversityBeijing 100091China school of engineering and computer science Victoria University of WellingtonWellington 6012New Zealand STEM University of South AustraliaAdelaide 5095Australia

Irregular boundaries in image stitching naturally occur due to freely moving *** deal with this problem,existing methods focus on optimizing mesh warping to make boundaries regular using the traditional explicit ***,previous methods always depend on hand-crafted features(e.g.,keypoints and line segments).Thus,failures often happen in overlapping regions without distinctive *** this paper,we address this problem by proposing RecStitchNet,a reasonable and effective network for image stitching with rectangular *** that both stitching and imposing rectangularity are non-trivial tasks in the learning-based framework,we propose a three-step progressive learning based strategy,which not only simplifies this task,but gradually achieves a good balance between stitching and imposing *** the first step,we perform initial stitching by a pre-trained state-of-the-art image stitching model,to produce initially warped stitching results without considering the boundary ***,we use a regression network with a comprehensive objective regarding mesh,perception,and shape to further encourage the stitched meshes to have rectangular boundaries with high content ***,we propose an unsupervised instance-wise optimization strategy to refine the stitched meshes iteratively,which can effectively improve the stitching results in terms of feature alignment,as well as boundary and structure *** to the lack of stitching datasets and the difficulty of label generation,we propose to generate a stitching dataset with rectangular stitched images as pseudo-ground-truth labels,and the performance upper bound induced from the it can be broken by our unsupervised *** and quantitative results and evaluations demonstrate the advantages of our method over the state-of-the-art.

关键词： image stitching boundaries convolutional neural network

Dynamic Dominating Set and Turbo-Charging Greedy Heuristics

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Tsinghua science and Technology 2014年第4期19卷 329-337页

作者： Rodney G.Downey Judith Egan Michael R.Fellows Frances A.Rosamond Peter Shaw school of Mathematics Statistics and Operations Research Victoria University of Wellington Wellington 600 New Zealand school of engineering and Information Technology Charles Darwin UniversityDarwin NT 0909 Australia

The main purpose of this paper is to exposit two very different, but very general, motivational schemes in the art of parameterization and a concrete example connecting them. We introduce a dynamic version of the DOMINATING SET problem and prove that it is fixed-parameter tractable（FPT）. The problem is motivated by settings where problem instances evolve. It also arises in the quest to improve a natural greedy heuristic for the DOMINATING SET problem.

关键词： kernelization multivariate algorithms parameterized algorithms turbo-charging heuristics