Filters








502 Hits in 10.2 sec

Deformable Generator Network: Unsupervised Disentanglement of Appearance and Geometry [article]

Xianglei Xing, Ruiqi Gao, Tian Han, Song-Chun Zhu, Ying Nian Wu
2020 arXiv   pre-print
We present a deformable generator model to disentangle the appearance and geometric information for both image and video data in a purely unsupervised manner.  ...  An extensive set of qualitative and quantitative experiments shows that the appearance and geometric information can be well disentangled, and the learned geometric generator can be conveniently transferred  ...  Our contributions are summarized below: • We propose a deformable generator network to disentangle the appearance and geometric information in a purely unsupervised manner.  ... 
arXiv:1806.06298v3 fatcat:cwx4l5crqjhsfagzctvoylvivq

Unsupervised Disentangling of Appearance and Geometry by Deformable Generator Network

Xianglei Xing, Tian Han, Ruiqi Gao, Song-Chun Zhu, Ying Nian Wu
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
We present a deformable generator model to disentangle the appearance and geometric information in purely unsupervised manner.  ...  An extensive set of qualitative and quantitative experiments shows that the appearance and geometric information can be well disentangled, and the learned geometric generator can be conveniently transferred  ...  Our contributions are summarized below: • Propose a deformable generator network to disentangle the appearance and geometric information in purely unsupervised manner. • The proposed method is general  ... 
doi:10.1109/cvpr.2019.01060 dblp:conf/cvpr/XingHGZW19 fatcat:q4shcfvdrnedtnxgn6ocsl2gde

DG-Font: Deformable Generative Networks for Unsupervised Font Generation [article]

Yangchen Xie and Xinyuan Chen and Li Sun and Yue Lu
2021 arXiv   pre-print
To address these problems, we propose novel deformable generative networks for unsupervised font generation (DGFont).  ...  Font generation is a challenging problem especially for some writing systems that consist of a large number of characters and has attracted a lot of attention in recent years.  ...  [52] disentangles image space into a Cartesian product of the appearance and the geometry latent spaces.  ... 
arXiv:2104.03064v2 fatcat:ho6dicd6cbdedg65qmetrh4uom

Disentangling Geometric Deformation Spaces in Generative Latent Shape Models [article]

Tristan Aumentado-Armstrong, Stavros Tsogkas, Sven Dickinson, Allan Jepson
2021 arXiv   pre-print
In this work, we improve on a prior generative model of geometric disentanglement for 3D shapes, wherein the space of object geometry is factorized into rigid orientation, non-rigid pose, and intrinsic  ...  Our improvements include more sophisticated handling of rotational invariance and the use of a diffeomorphic flow network to bridge latent and spectral space.  ...  Our method relies on the isometry invariance of the LBOS, which can be estimated from the geometry directly, and uses disentanglement techniques to partition the latent space of a generative model into  ... 
arXiv:2103.00142v1 fatcat:oefxuozl5bewtcnbvqoe66ccze

Deforming Autoencoders: Unsupervised Disentangling of Shape and Appearance [chapter]

Zhixin Shu, Mihir Sahasrabudhe, Rıza Alp Güler, Dimitris Samaras, Nikos Paragios, Iasonas Kokkinos
2018 Lecture Notes in Computer Science  
In this work we introduce Deforming Autoencoders, a generative model for images that disentangles shape from appearance in an unsupervised manner.  ...  We also achieve a more powerful form of unsupervised disentangling in template coordinates, that successfully decomposes face images into shading and albedo, allowing us to further manipulate face images  ...  Acknowledgment This work was supported by a gift from Adobe, NSF grants CNS-1718014 and DMS 1737876, the Partner University Fund, and the SUNY2020 Infrastructure Transportation Security Center.  ... 
doi:10.1007/978-3-030-01249-6_40 fatcat:k4c62he67bgybail7in6w7x35e

Region Deformer Networks for Unsupervised Depth Estimation from Unconstrained Monocular Videos [article]

Haofei Xu, Jianmin Zheng, Jianfei Cai, Juyong Zhang
2019 arXiv   pre-print
In this paper, we propose a new learning based method consisting of DepthNet, PoseNet and Region Deformer Networks (RDN) to estimate depth from unconstrained monocular videos without ground truth supervision  ...  The core contribution lies in RDN for proper handling of rigid and non-rigid motions of various objects such as rigidly moving cars and deformable humans.  ...  STN and Deformable ConvNets are both aiming at designing network architectures with geometry invariant for supervised tasks like classification and segmentation.  ... 
arXiv:1902.09907v2 fatcat:4jfeqpptanghnhkjkq5hojrod4

Deforming Autoencoders: Unsupervised Disentangling of Shape and Appearance [article]

Zhixin Shu, Mihir Sahasrabudhe, Alp Guler, Dimitris Samaras, Nikos Paragios, Iasonas Kokkinos
2018 arXiv   pre-print
In this work we introduce Deforming Autoencoders, a generative model for images that disentangles shape from appearance in an unsupervised manner.  ...  A more powerful form of unsupervised disentangling becomes possible in template coordinates, allowing us to successfully decompose face images into shading and albedo, and further manipulate face images  ...  Using these datasets we experimentally explored the ability of the unsupervised appearance-shape (or texture-deformation) disentangling network on 1) unsupervised image alignment/appearance inference;  ... 
arXiv:1806.06503v1 fatcat:2y3w7ofn6fhzrkac27gsabrg74

MoCaNet: Motion Retargeting In-the-Wild via Canonicalization Networks

Wentao Zhu, Zhuoqian Yang, Ziang Di, Wayne Wu, Yizhou Wang, Chen Change Loy
2022 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Notably, the canonicalized skeleton sequence could serve as a disentangled and interpretable representation of human motion that benefits action analysis and motion retrieval.  ...  It is designed to leverage massive online videos for unsupervised training, needless of 3D annotations or motion-body pairing information.  ...  We thank Xiaoxuan Ma, Shikai Li, Jiajun Su and Peizhuo Li for insightful discussion and kind support.  ... 
doi:10.1609/aaai.v36i3.20274 fatcat:lidadulcffgtppjcxbxbgfhfqy

Lifting AutoEncoders: Unsupervised Learning of a Fully-Disentangled 3D Morphable Model Using Deep Non-Rigid Structure From Motion

Mihir Sahasrabudhe, Zhixin Shu, Edward Bartrum, Riza Alp Guler, Dimitris Samaras, Iasonas Kokkinos
2019 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)  
The learned generative model can be controlled in terms of interpretable geometry and appearance factors, allowing us to perform photorealistic image manipulation of identity, expression, 3D pose, and  ...  We exploit the 3D geometric nature of our model and use normal information to disentangle appearance into illumination, shading, and albedo.  ...  This work was supported by a gift from Adobe, NSF grants CNS-1718014 and DMS 1737876, the Partner University Fund, and the SUNY2020 Infrastructure Transportation Security Center.  ... 
doi:10.1109/iccvw.2019.00500 dblp:conf/iccvw/SahasrabudheSBG19 fatcat:f3dv3unkqnhvphph76laefew54

Lifting AutoEncoders: Unsupervised Learning of a Fully-Disentangled 3D Morphable Model using Deep Non-Rigid Structure from Motion [article]

Mihir Sahasrabudhe, Zhixin Shu, Edward Bartrum, Riza Alp Guler, Dimitris Samaras, Iasonas Kokkinos
2019 arXiv   pre-print
The learned generative model can be controlled in terms of interpretable geometry and appearance factors, allowing us to perform photorealistic image manipulation of identity, expression, 3D pose, and  ...  We exploit the 3D geometric nature of our model and use normal information to disentangle appearance into illumination, shading and albedo.  ...  geometry of a deformable object category in an entirely unsupervised manner from an unstructured collection of RGB images.  ... 
arXiv:1904.11960v1 fatcat:ohpo2bu3rbe2hm5dwq4k55cnv4

MoCaNet: Motion Retargeting in-the-wild via Canonicalization Networks [article]

Wentao Zhu, Zhuoqian Yang, Ziang Di, Wayne Wu, Yizhou Wang, Chen Change Loy
2021 arXiv   pre-print
Notably, the canonicalized skeleton sequence could serve as a disentangled and interpretable representation of human motion that benefits action analysis and motion retrieval.  ...  It is designed to leverage massive online videos for unsupervised training, needless of 3D annotations or motion-body pairing information.  ...  We thank Xiaoxuan Ma, Shikai Li, Jiajun Su and Peizhuo Li for insightful discussion and kind support.  ... 
arXiv:2112.10082v2 fatcat:ccpfnww44rcxpdeiyejklhxl7q

Disentangled Human Body Embedding Based on Deep Hierarchical Neural Network [article]

Boyi Jiang, Juyong Zhang, Jianfei Cai, Jianmin Zheng
2020 IEEE Transactions on Visualization and Computer Graphics   accepted
the learning of the neural network.  ...  This paper presents an autoencoder-like network architecture to learn disentangled shape and pose embedding specifically for the 3D human body.  ...  [23] proposed a novel strategy to automatically learn disentangled latent representations from raw data in a completely unsupervised manner. Deformation representation.  ... 
doi:10.1109/tvcg.2020.2988476 pmid:32324557 arXiv:1905.05622v2 fatcat:tmdgr54fnjdxlk2ukburvkivxy

Unsupervised Learning of Object Landmarks through Conditional Image Generation [article]

Tomas Jakab, Ankush Gupta, Hakan Bilen, Andrea Vedaldi
2018 arXiv   pre-print
Compared to standard image generation problems, which often use generative adversarial networks, our generation task is conditioned on both appearance and geometry and thus is significantly less ambiguous  ...  We cast this as the problem of generating images that combine the appearance of the object as seen in a first example image with the geometry of the object as seen in a second example image, where the  ...  We would like to thank James Thewlis for suggestions and support with code and data, and David Novotný and Triantafyllos Afouras for helpful advice.  ... 
arXiv:1806.07823v2 fatcat:wiypxze42vbbfm6pib6rgtcqwq

A State-of-the-Art Review on Image Synthesis with Generative Adversarial Networks

Lei Wang, Wei Chen, Wenjia Yang, Fangming Bi, Fei Richard Yu
2020 IEEE Access  
Generative Adversarial Networks (GANs) have achieved impressive results in various image synthesis tasks, and are becoming a hot topic in computer vision research because of the impressive performance  ...  INDEX TERMS Generative adversarial networks, image synthesis, image-to-image translation, image editing, cartoon generation.  ...  It learns a translation which is built on appearance and geometry space separately by disentangling the image space into an appearance space and a geometry latent space to decompose image-to-image translation  ... 
doi:10.1109/access.2020.2982224 fatcat:p5uxjh4cybfw5grp6ldhkpukrm

Flow Guided Transformable Bottleneck Networks for Motion Retargeting [article]

Jian Ren, Menglei Chai, Oliver J. Woodford, Kyle Olszewski, Sergey Tulyakov
2021 arXiv   pre-print
Inspired by the Transformable Bottleneck Network, which renders novel views and manipulations of rigid objects, we propose an approach based on an implicit volumetric representation of the image content  ...  However, the scalability of such methods is limited, as each model can only generate videos for the given target subject, and such training videos are labor-intensive to acquire and process.  ...  [37] show that body parts can also be represented by unsupervised learning, which also helps to disentangle body pose and shape [11] .  ... 
arXiv:2106.07771v1 fatcat:yjprgqyphbdenkfxguq5dpysxu
« Previous Showing results 1 — 15 out of 502 results