Filters








6,320 Hits in 3.8 sec

Surface Reconstruction with Data-driven Exemplar Priors [article]

Oussama Remil, Qian Xie, Xingyu Xie, Kai Xu, Jun Wang
2017 arXiv   pre-print
By manipulating these priors, we are able to reconstruct geometrically faithful models with the same class of objects from raw point clouds.  ...  We then formulate the problem to select as-few-as-possible priors from the library, referred to as exemplar priors.  ...  (a) Training dataset with a set of learned exemplar priors. (b) Input point data with 10% Gaussian noise and the reconstruction result using RIMLS.  ... 
arXiv:1701.03230v1 fatcat:qv23qbb46vchvnpp6uqaqw4smm

Data-driven assimilation of irregularly-sampled image time series

R. Fablet, P. Viet, R. Lguensat, B. Chapron
2017 2017 IEEE International Conference on Image Processing (ICIP)  
We address in this paper the reconstruction of irregurlarlysampled image time series with an emphasis on geophysical remote sensing data.  ...  Contrary to model-driven assimilation models, we do not exploit a physically-derived dynamic prior but we build a data-driven dynamic prior from a representative dataset of the considered image dynamics  ...  Fig.1 illustrates the scarcity of the data for such applications, here a satellite-derived sea surface geophysical field with missing data rates greater than 90% for a particular day and area.  ... 
doi:10.1109/icip.2017.8297094 dblp:conf/icip/FabletVLC17 fatcat:ksixkt6zxrfdfinfmyjbydj4je

Data-Driven Flower Petal Modeling with Botany Priors

Chenxi Zhang, Mao Ye, Bo Fu, Ruigang Yang
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
We overcome these challenges by combining data driven modeling techniques with domain knowledge from botany.  ...  Each segmented petal is then fitted with a scale-invariant morphable petal shape model, which is constructed from individually scanned exemplar petals.  ...  Our method uses both a data-driven approach and knowledge in botany to handle these challenges.  ... 
doi:10.1109/cvpr.2014.87 dblp:conf/cvpr/ZhangYFY14 fatcat:d3akzk675nhjdk74dnropkjv6m

Data-Driven Shape Analysis and Processing [article]

Kai Xu, Vladimir G. Kim, Qixing Huang, Evangelos Kalogerakis
2015 arXiv   pre-print
We conclude our report with ideas that can inspire future research in data-driven shape analysis and processing.  ...  Data-driven methods play an increasingly important role in discovering geometric, structural, and semantic relationships between 3D shapes in collections, and applying this analysis to support intelligent  ...  Traditional shape reconstruction algorithms usually utilize generic priors, such as surface smoothness [DTB06] , and typically assume that the input data captures most of the object's surface.  ... 
arXiv:1502.06686v1 fatcat:upajios4y5a6dgf2zw7faqai4a

Joint Interpolation of Multisensor Sea Surface Temperature Fields Using Nonlocal and Statistical Priors

Ronan Fablet, Francois Rousseau
2016 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
The exemplar-based prior, referred to as a non-local prior, exploits similarities between local patches (small field regions) to interpolate missing data areas from previously observed exemplars.  ...  We propose a novel statistical model, which combines two key features: an exemplar-based prior and second-order statistical priors.  ...  Piolle from Ifremer/CERSAT for providing the SST data.  ... 
doi:10.1109/jstars.2016.2523605 fatcat:zeycn6jxjjdgfhrpe4jusyq764

Data-Driven Shape Analysis and Processing

Kai Xu, Vladimir G. Kim, Qixing Huang, Evangelos Kalogerakis
2016 Computer graphics forum (Print)  
We conclude our report with ideas that can inspire future research in data-driven shape analysis and processing.  ...  Data-driven methods serve an increasingly important role in discovering geometric, structural and semantic relationships between shapes.  ...  Traditional shape reconstruction algorithms usually utilize generic priors, such as surface smoothness [DTB06] , and typically assume that the input data capture most of the object's surface.  ... 
doi:10.1111/cgf.12790 fatcat:q76sq2syjvce5fjjjb7yoafww4

A Survey of Surface Reconstruction from Point Clouds

Matthew Berger, Andrea Tagliasacchi, Lee M. Seversky, Pierre Alliez, Gaël Guennebaud, Joshua A. Levine, Andrei Sharf, Claudio T. Silva
2016 Computer graphics forum (Print)  
We survey the field of surface reconstruction, and provide a categorization with respect to priors, data imperfections, and reconstruction output.  ...  future work in surface reconstruction.  ...  User Driven The user driven prior incorporates the user into the process of surface reconstruction, allowing for them to provide intuitive and useful cues for reconstruction.  ... 
doi:10.1111/cgf.12802 fatcat:r4qj7emtpzfphajc2hgbqgampi

Analysis and Modeling of 3D Indoor Scenes [article]

Rui Ma
2017 arXiv   pre-print
With the fast development of VR/AR devices and the emerging applications, the demand of realistic 3D indoor scenes keeps growing rapidly.  ...  We first review works on understanding and semantic modeling of scenes from captured 3D data of the real world.  ...  The key for data-driven approaches is the source of data, especially those with rich and semantic annotations.  ... 
arXiv:1706.09577v1 fatcat:pdbztyjkezabnj3eaok5wrs7xq

Unsupervised Reconstruction of Sea Surface Currents from AIS Maritime Traffic Data Using Trainable Variational Models

Simon Benaïchouche, Clément Legoff, Yann Guichoux, François Rousseau, Ronan Fablet
2021 Remote Sensing  
They support the relevance of the proposed learning-based AIS-driven approach to significantly improve the reconstruction of sea surface currents compared with state-of-the-art methods, including altimetry-based  ...  The latter exploits variational auto-encoders coupled with neural ODE to represent sea surface dynamics.  ...  Given a trained model with parameters Θ, we may solve minimization (2) to reconstruct the sea surface currents associated with any new AIS dataset.  ... 
doi:10.3390/rs13163162 fatcat:z4l5avlamjgordwtjdxekjaqlm

Data-driven shape analysis and processing

Kai Xu, Vladimir G. Kim, Qixing Huang, Niloy Mitra, Evangelos Kalogerakis
2016 SIGGRAPH ASIA 2016 Courses on - SA '16  
Prior to the emergence of data-driven techniques, high-level shape understanding and modeling was usually achieved with knowledge-driven methods.  ...  We conclude our report with ideas that can inspire future research in data-driven shape analysis and processing. Relation to knowledge-driven shape processing.  ...  Traditional shape reconstruction algorithms usually utilize generic priors, such as surface smoothness [Diebel et al. 2006] , and typically assume that the input data captures most of the object's surface  ... 
doi:10.1145/2988458.2988473 dblp:conf/siggraph/0004KHMK16 fatcat:tefja76ijnclzmpux2iaj45zgu

Guided Fine-Tuning for Large-Scale Material Transfer [article]

Valentin Deschaintre, George Drettakis, Adrien Bousseau
2020 arXiv   pre-print
We use existing methods to extract SVBRDF parameters from the close-ups, and our method to transfer these parameters to the entire surface, enabling the lightweight capture of surfaces several meters wide  ...  We present a method to transfer the appearance of one or a few exemplar SVBRDFs to a target image representing similar materials.  ...  Our method better transfers details of the surface compared to prior work, which either only captures global statistics (1st row) or struggles to generalize from a limited exemplar (2nd row). Method  ... 
arXiv:2007.03059v1 fatcat:xfkrb75brjhjvorwbtf7eiyc2u

BAE-NET: Branched Autoencoder for Shape Co-Segmentation [article]

Zhiqin Chen, Kangxue Yin, Matthew Fisher, Siddhartha Chaudhuri, Hao Zhang
2019 arXiv   pre-print
By complementing the shape reconstruction loss with a label loss, BAE-NET is easily tuned for one-shot learning.  ...  The unsupervised BAE-NET is trained with a collection of un-segmented shapes, using a shape reconstruction loss, without any ground-truth labels.  ...  Many prior unsupervised co-segmentation methods [12, 18, 20, 50] , which are model-driven rather than data-driven, had only been tested on very small input sets (less than 50 shapes).  ... 
arXiv:1903.11228v2 fatcat:htxwsw6wovdjzmzrb2hiveb6zi

Extended Patch Prioritization for Depth Filling Within Constrained Exemplar-Based RGB-D Image Completion [chapter]

Amir Atapour-Abarghouei, Toby P. Breckon
2018 Lecture Notes in Computer Science  
Furthermore, the query space in the source region is constrained to increase the efficiency of the approach compared to other exemplar-driven methods.  ...  Most existing exemplar-based inpainting techniques, designed for color image completion, do not perform well on depth information with object boundaries obstructed or surrounded by missing regions.  ...  Prior Work Prior work in depth hole filling [7, 9, 12, 14, 23, 24] is not as comprehensive as color image completion.  ... 
doi:10.1007/978-3-319-93000-8_35 fatcat:ymasx4jcpvdube2zhsp7pfm6fe

Morfit

Kangxue Yin, Hui Huang, Hao Zhang, Minglun Gong, Daniel Cohen-Or, Baoquan Chen
2014 ACM Transactions on Graphics  
With significant data missing in a point scan, reconstructing a complete surface with sufficient geometric and topological fidelity is highly challenging.  ...  We demonstrate various interactive reconstructions from point scans with sharp features and significant missing data.  ...  As we demonstrate, our curve-driven interactive reconstruction technique leads to successful interpolation of surfaces over regions with significant missing data, sharp features, and general geometric  ... 
doi:10.1145/2661229.2661241 fatcat:obw7enc6m5erpidvplfcmss7te

Spatio-Temporal Interpolation of Cloudy SST Fields Using Conditional Analog Data Assimilation

Ronan Fablet, Phi Viet, Redouane Lguensat, Pierre-Henri Horrein, Bertrand Chapron
2018 Remote Sensing  
Physically-driven data assimilation models [20] may outperform OI if relevant dynamical priors can be defined [21] .  ...  EOF-based models were among the early and perhaps most popular data-driven methods applied to the reconstruction of SST fields from cloudy SST data [14, 15, 22] as well of other sea surface tracers such  ...  P.V. and R.L. analyzed the data; P.V., R.L. and PH.H. contributed reagents/materials/analysis tools; R.F. wrote the paper."  ... 
doi:10.3390/rs10020310 fatcat:zylrpczzg5evfdviugtvzgdpua
« Previous Showing results 1 — 15 out of 6,320 results