A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf
.
Filters
Generative Visual Manipulation on the Natural Image Manifold
[article]
2018
arXiv
pre-print
The presented method can further be used for changing one image to look like the other, as well as generating novel imagery from scratch based on user's scribbles. ...
All our manipulations are expressed in terms of constrained optimization and are applied in near-real time. We evaluate our algorithm on the task of realistic photo manipulation of shape and color. ...
Though lacking visual details sometimes, the model can synthesize appealing samples with a plausible overall structure. ...
arXiv:1609.03552v3
fatcat:4p4nhm2uhjbpdfzekiijr7utfa
Generative Visual Manipulation on the Natural Image Manifold
[chapter]
2016
Lecture Notes in Computer Science
The presented method can further be used for changing one image to look like the other, as well as generating novel imagery from scratch based on user's scribbles. ...
All our manipulations are expressed in terms of constrained optimization and are applied in near-real time. We evaluate our algorithm on the task of realistic photo manipulation of shape and color. ...
Though lacking visual details sometimes, the model can synthesize appealing samples with a plausible overall structure. ...
doi:10.1007/978-3-319-46454-1_36
fatcat:y64u4pnkxng3jnlogi7j2p443y
One-Shot Generation of Near-Optimal Topology through Theory-Driven Machine Learning
[article]
2018
arXiv
pre-print
Deviation of the student's solutions from the optimality conditions is quantified, and used for choosing new data points to learn from. ...
We show through a compliance minimization problem that the proposed learning mechanism leads to topology generation with near-optimal structural compliance, much improved from standard supervised learning ...
The rest of the paper is structured as follows: In Sec. 2 we review related work at the intersection of generative design and machine learning, and highlight the new contributions ...
arXiv:1807.10787v3
fatcat:bwhnr5epmnemzd3gjlysnzcshu
Example-Based Human Motion Extrapolation and Motion Repairing Using Contour Manifold
2014
IEEE transactions on multimedia
Contour manifold construction searches for low-dimensional manifolds that represent the temporal-domain deformation of the reference motion sequence. ...
The algorithm is implemented in two major steps: contour manifold construction and object motion synthesis. ...
In [5] , Fang and Pollard used a validity constraint optimization approach that iteratively adjusts a synthesized motion to satisfy the animator's requirement for realistic motions. ...
doi:10.1109/tmm.2013.2283844
fatcat:ofvuwc6iovdtpb55lfm66d3itq
Manifold Criterion Guided Transfer Learning via Intermediate Domain Generation
[article]
2019
arXiv
pre-print
Experiments on a number of benchmark visual transfer tasks demonstrate the superiority of the proposed manifold criterion guided generative transfer method, by comparing with other state-of-the-art methods ...
For better exploiting the domain locality, a novel local generative discrepancy metric (LGDM) based intermediate domain generation learning called Manifold Criterion guided Transfer Learning (MCTL) is ...
ACKNOWLEDGMENT The authors would like to thank the editor and the anonymous reviewers for their valuable comments and suggestions. ...
arXiv:1903.10211v1
fatcat:yv5eeuggonbsbbs5v7ulswflxi
Fast Neural Style Transfer for Motion Data
2017
IEEE Computer Graphics and Applications
For tasks such as style transfer this data may not always be available and so a different training method is required. ...
We present a fast, efficient technique for performing neural style transfer of human motion data using a feedforward neural network. ...
They then provide a framework to edit the generated motions using the motion manifold and by optimizing the motion in the hidden unit space to satisfy constraints such as bone-length and foot sliding. ...
doi:10.1109/mcg.2017.3271464
pmid:28829292
fatcat:4wdfbgfe65htfmvd4fwp5iwfru
A unified shape editing framework based on tetrahedral control mesh
2009
Computer Animation and Virtual Worlds
And an error-driven refinement approach is presented to further improve the deformation result. ...
Experimental results show our algorithm is effective, easy to control, supports various shape representations, and well transfers deformations between non-homeomorphous models. ...
TO: time for transfer optimization and modified-BI. ...
doi:10.1002/cav.302
fatcat:4fnjlaum4vbupclfzqlhz5nknq
Neuro-Visualizer: An Auto-encoder-based Loss Landscape Visualization Method
[article]
2023
arXiv
pre-print
In this paper, we present a novel auto-encoder-based non-linear landscape visualization method called Neuro-Visualizer that addresses these shortcoming and provides useful insights about neural network ...
In recent years, there has been a growing interest in visualizing the loss landscape of neural networks. ...
mapping such manifolds onto a 2-D grid for loss landscape visualization. ...
arXiv:2309.14601v1
fatcat:l664vrqhwbawtljv6iaabidhz4
Procedural Editing of Bidirectional Texture Functions
[article]
2007
Symposium on Rendering
all visually relevant details. ...
It is based on the observation that we are already good in modeling the basic geometric structure of many natural and manmade materials but still have not found effective models for the detailed small-scale ...
If the constraint differs significantly from the reconstructed meso-structure it can happen that visually important structures brake up and are not correctly transferred especially for non-frontal viewing ...
doi:10.2312/egwr/egsr07/219-230
fatcat:vxmaq3rrgfeppg6bgtvasbvaae
Unsupervised Topological Alignment for Single-Cell Multi-Omics Integration
[article]
2020
bioRxiv
pre-print
by matching the distance matrices via a matrix optimization method. ...
structures. ...
For complex embedded hierarchical structures with multi-scales, UnionCom can align the manifold recursively by introducing scaling-specific factors for each scale of the manifold, and we plan to pursue ...
doi:10.1101/2020.02.02.931394
fatcat:imqynkxa5badnf2ax2zorbtjfa
A Framework for Data-Driven Computational Mechanics Based on Nonlinear Optimization
[article]
2019
arXiv
pre-print
The second one is a data driven inverse approach that seeks to reconstruct a constitutive manifold from data sets by manifold learning techniques, relying on a well-defined functional structure of the ...
We discuss important mathematical aspects of our approach for a data-driven truss element and investigate its key numerical behavior for a data-driven beam element that makes use of all components of our ...
Conclusions In this work, we presented an approximate nonlinear optimization problem for Data-Driven Computational Mechanics that enables us to handle: i) kinematic constraints; and, ii) materials whose ...
arXiv:1910.12736v1
fatcat:fknlgtpcijeujmqmhlruwjw7pe
A Revisit of Shape Editing Techniques: from the Geometric to the Neural Viewpoint
[article]
2021
arXiv
pre-print
Traditionally, the deformed shape is determined by the optimal transformation and weights for an energy term. ...
With increasing availability of 3D shapes on the Internet, data-driven methods were proposed to improve the editing results. ...
The deformation results using optimized weights can better reflect the deformation principle of the example shapes in the dataset.The above methods are suitable for manifold meshes. ...
arXiv:2103.01694v1
fatcat:lhgswnemnbhvrazl76qhz5rhmy
Cyclic Functional Mapping: Self-supervised correspondence between non-isometric deformable shapes
[article]
2019
arXiv
pre-print
We present the first utterly self-supervised network for dense correspondence mapping between non-isometric shapes. ...
As the same learnable rules that generate the point-wise descriptors apply in both directions, the network learns invariant structures without any labels while coping with non-isometric deformations. ...
transfer. ...
arXiv:1912.01249v1
fatcat:gpx7lflrlzgmxfkg4bbzdxfh6e
Data-driven modeling and learning in science and engineering
2019
Comptes rendus. Mecanique
Some scientific fields have been using artificial intelligence for some time due to the inherent difficulty in obtaining laws and equations to describe some phenomena. ...
Data-driven modeling and scientific discovery is a change of paradigm on how many problems, both in science and engineering, are addressed. ...
Data-driven dimension reduction procedures applied to dynamical systems, both for modal decompositions and for transfer functions, are studied in [99] among others. ...
doi:10.1016/j.crme.2019.11.009
fatcat:7rtlth7ncreqthugtduxtzjpky
Author Index
2010
2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Connectivity Constraints for Reconstruction of 3D Line Segments from Images Tommasi, Tatiana Safety in Numbers: Learning Categories from Few Examples with Multi Model Knowledge Transfer Tong, Yan Workshop ...
Boundary Learning by Optimization with Topological Constraints
He, Kaiming
Fast Matting Using Large Kernel Matting Laplacian Matrices
He, Lei
Object Matching with a Locally Affine-Invariant Constraint ...
doi:10.1109/cvpr.2010.5539913
fatcat:y6m5knstrzfyfin6jzusc42p54
« Previous
Showing results 1 — 15 out of 11,506 results