833 Hits in 5.7 sec

Recurrent Semi-supervised Classification and Constrained Adversarial Generation with Motion Capture Data [article]

Félix G. Harvey, Julien Roy, David Kanaa, Christopher Pal
2018 arXiv   pre-print
We explore recurrent encoder multi-decoder neural network architectures for semi-supervised sequence classification and reconstruction.  ...  We further explore a novel formulation for future-predicting decoders based on conditional recurrent generative adversarial networks, for which we propose both soft and hard constraints for transition  ...  Acknowledgements We thank Ubisoft and the Natural Sciences and Engineering Research Council of Canada for support under the Collaborative Research and Development program.  ... 
arXiv:1511.06653v8 fatcat:q3xh3aelfzegpcza4wj3rqw2q4


Lilang Lin, Sijie Song, Wenhan Yang, Jiaying Liu
2020 Proceedings of the 28th ACM International Conference on Multimedia  
We evaluate our multi-task self-supervised learning approach with action classifiers trained under different configurations, including unsupervised, semi-supervised and fully-supervised settings.  ...  Instead, we propose to integrate multiple tasks to learn more general representations in a self-supervised manner.  ...  In semi-supervised learning, the training process utilizes both labeled data and unlabeled data.  ... 
doi:10.1145/3394171.3413548 dblp:conf/mm/LinSY020 fatcat:vgmk7qtc7vfrbgktb2ae44sck4

GAN-based Reactive Motion Synthesis with Class-aware Discriminators for Human-human Interaction [article]

Qianhui Men, Hubert P. H. Shum, Edmond S. L. Ho, Howard Leung
2021 arXiv   pre-print
This allows the use of such labels in supervising the training of the generator. We experiment with the SBU and the HHOI datasets.  ...  In this paper, we propose a semi-supervised GAN system that synthesizes the reactive motion of a character given the active motion from another character. Our key insights are two-fold.  ...  This follows the idea of semi-supervised learning with GAN from [22, 23] , where they generate semi-supervised generative framework with an unsupervised discriminator to tell the fidelity of the generation  ... 
arXiv:2110.00380v1 fatcat:ew6e2oynuzbexopchgreqny3em

Adversarial Framework for Unsupervised Learning of Motion Dynamics in Videos [article]

C. Spampinato, S. Palazzo, P. D'Oro, D. Giordano, M. Shah
2019 arXiv   pre-print
Self-supervision is enforced by using motion masks produced by the generator, as a co-product of its generation process, to supervise the discriminator network in performing dense prediction.  ...  Current approaches based on supervised learning require large amounts of annotated data, whose scarce availability is one of the main limiting factors to the development of general solutions.  ...  The network is trained both on labeled and unlabeled data and the adversarial GAN loss is extended with the supervised end-point-error loss, computed on the labeled data.  ... 
arXiv:1803.09092v2 fatcat:tconl7knq5af3nqlbxthvx7br4

Through-Wall Human Mesh Recovery Using Radio Signals

Mingmin Zhao, Yingcheng Liu, Aniruddh Raghu, Hang Zhao, Tianhong Li, Antonio Torralba, Dina Katabi
2019 2019 IEEE/CVF International Conference on Computer Vision (ICCV)  
(b) shows the dynamic meshes that capture the motion when the person walks, waves his hand, and sits.  ...  Images captured by a camera co-located with the radio sensor are presented here for visual reference.  ...  Capturing the prior of human shape and human motion dynamics is essential in order to generate accurate and realistic dynamic meshes.  ... 
doi:10.1109/iccv.2019.01021 dblp:conf/iccv/ZhaoLRZL0K19 fatcat:d3qrriyv3vgjbguams5atoge2e

2020 Index IEEE Transactions on Image Processing Vol. 29

2020 IEEE Transactions on Image Processing  
., +, TIP 2020 2066-2077 Semi-Supervised Deep Coupled Ensemble Learning With Classification Landmark Exploration.  ...  ., +, TIP 2020 8606-8621 Data privacy PrivacyNet: Semi-Adversarial Networks for Multi-Attribute Face Privacy.  ... 
doi:10.1109/tip.2020.3046056 fatcat:24m6k2elprf2nfmucbjzhvzk3m

2021 Index IEEE Transactions on Image Processing Vol. 30

2021 IEEE Transactions on Image Processing  
The primary entry includes the coauthors' names, the title of the paper or other item, and its location, specified by the publication abbreviation, year, month, and inclusive pagination.  ...  The Subject Index contains entries describing the item under all appropriate subject headings, plus the first author's name, the publication abbreviation, month, and year, and inclusive pages.  ...  Direct Unsupervised Super-Resolution Using Generative Adversarial Network (DUS-GAN) for Real-World Data.  ... 
doi:10.1109/tip.2022.3142569 fatcat:z26yhwuecbgrnb2czhwjlf73qu

Anomalous Human Activity Recognition in Surveillance Videos

2019 International journal of recent technology and engineering  
The objective of the discussion is to be able to implement an automated anomalous human activity recognition system which uses surveillance video to capture the occurrence of an unusual event and caution  ...  Accurate human activity recognition in real-time is challenging because human activities are complicated and extremely diverse in nature.  ...  Shape and motion templates are the result of classification of the pose template. It is emphasized on SVM (Support Vector Machine) algorithm which is built on the semi-supervised structure.  ... 
doi:10.35940/ijrte.b1064.0782s719 fatcat:352q6ou655dr7oj2jtk2rl6eca

Deep Weakly-Supervised Domain Adaptation for Pain Localization in Videos [article]

Gnana Praveen R, Eric Granger, Patrick Cardinal
2020 arXiv   pre-print
conditions, and lack of representative training videos with labels.  ...  The training process relies on weak target loss, along with domain loss and source loss for domain adaptation of the I3D model.  ...  the domain classification loss to constrain the features to be domain-invariant.  ... 
arXiv:1910.08173v2 fatcat:p7huseb36ve37nk4qpmbygnazq

Instance-Level Future Motion Estimation in a Single Image Based on Ordinal Regression and Semi-Supervised Domain Adaptation

Kyung-Rae Kim, Yeong Jun Koh, Chang-Su Kim
2020 IEEE Access  
INDEX TERMS Future motion estimation, cyclic ordinal regression, semi-supervised domain adaptation.  ...  A novel algorithm to estimate instance-level future motion (FM) in a single image is proposed in this paper. First, the FM of an instance is defined with its direction, speed, and action classes.  ...  An adversarial training method is developed to perform semi-supervised domain adaptation using three kinds of data: a large number of labeled source domain data, only a limited number of labeled data in  ... 
doi:10.1109/access.2020.3003751 fatcat:2fdmo227x5dtfhctk7g3k3qw3i

Self-supervised Visual Feature Learning with Deep Neural Networks: A Survey [article]

Longlong Jing, Yingli Tian
2019 arXiv   pre-print
To avoid extensive cost of collecting and annotating large-scale datasets, as a subset of unsupervised learning methods, self-supervised learning methods are proposed to learn general image and video features  ...  First, the motivation, general pipeline, and terminologies of this field are described. Then the common deep neural network architectures that used for self-supervised learning are summarized.  ...  The adversarial training can help the network to capture the real distribution of the real data and generate realists data, and it has been widely used in computer vision tasks such as image generation  ... 
arXiv:1902.06162v1 fatcat:wwc3nenj3vbybcrd7gx2jytlte

Complex Deep Neural Networks from Large Scale Virtual IMU Data for Effective Human Activity Recognition Using Wearables

Hyeokhyen Kwon, Gregory D. Abowd, Thomas Plötz
2021 Sensors  
The proposed model is trained with the large amount of virtual IMU data and calibrated with a mere 36 min of real IMU data.  ...  Supervised training of human activity recognition (HAR) systems based on body-worn inertial measurement units (IMUs) is often constrained by the typically rather small amounts of labeled sample data.  ...  The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results  ... 
doi:10.3390/s21248337 pmid:34960431 pmcid:PMC8707382 fatcat:ex27jvnfxnbq3g7b2wfdqmmy2a

2020 Index IEEE Transactions on Knowledge and Data Engineering Vol. 32

2021 IEEE Transactions on Knowledge and Data Engineering  
., +, TKDE Oct. 2020 1881-1896 Classification Capturing Joint Label Distribution for Multi-Label Classification Through Adversarial Learning.  ...  ., +, TKDE Jan. 2020 165-176 Semi-Supervised Learning with Auto-Weighting Feature and Adaptive Graph.  ... 
doi:10.1109/tkde.2020.3038549 fatcat:75f5fmdrpjcwrasjylewyivtmu

Monocular Depth Estimation Based On Deep Learning: An Overview [article]

Chaoqiang Zhao, Qiyu Sun, Chongzhen Zhang, Yang Tang, Feng Qian
2020 arXiv   pre-print
Furthermore, we review some representative existing methods according to different training manners: supervised, unsupervised and semi-supervised.  ...  Traditional depth estimation methods, like structure from motion and stereo vision matching, are built on feature correspondences of multiple viewpoints.  ...  Different kinds of adversarial [62] has the ability to generate the data from a vector z, and the discriminator is designed to distinguish the real and fake data.  ... 
arXiv:2003.06620v1 fatcat:l5ei3ognova6xkyppflef5nqsq

2021 Index IEEE Transactions on Multimedia Vol. 23

2021 IEEE transactions on multimedia  
The primary entry includes the coauthors' names, the title of the paper or other item, and its location, specified by the publication abbreviation, year, month, and inclusive pagination.  ...  The Subject Index contains entries describing the item under all appropriate subject headings, plus the first author's name, the publication abbreviation, month, and year, and inclusive pages.  ...  RealVAD: A Real-World Dataset and A Method for Voice Activity Detection by Body Motion Analysis. Beyan, C., +, TMM 2021 2071-2085 Recurrent Generative Adversarial Network for Face Completion.  ... 
doi:10.1109/tmm.2022.3141947 fatcat:lil2nf3vd5ehbfgtslulu7y3lq
« Previous Showing results 1 — 15 out of 833 results