Filters








67,026 Hits in 4.0 sec

Unsupervised Learning of Edges

Yin Li, Manohar Paluri, James M. Rehg, Piotr Dollar
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Is this form of strong, highlevel supervision actually necessary to learn to accurately detect edges?  ...  Using a large corpus of video data, we show that edge detectors trained using our unsupervised scheme approach the performance of the same methods trained with full supervision (within 3-5%).  ...  Yin Li gratefully acknowledges the support of the Intel ISTC-PC while completing the writing of the paper at Georgia Tech.  ... 
doi:10.1109/cvpr.2016.179 dblp:conf/cvpr/LiPRD16 fatcat:ltclg336tbey3p4va4wq2gwqma

Unsupervised Learning of Edges [article]

Yin Li and Manohar Paluri and James M. Rehg and Piotr Dollár
2016 arXiv   pre-print
Is this form of strong, high-level supervision actually necessary to learn to accurately detect edges?  ...  Using a large corpus of video data, we show that edge detectors trained using our unsupervised scheme approach the performance of the same methods trained with full supervision (within 3-5%).  ...  Yin Li gratefully acknowledges the support of the Intel ISTC-PC while completing the writing of the paper at Georgia Tech.  ... 
arXiv:1511.04166v2 fatcat:7aqvjj2gofgm5cjgb3yxxgd4y4

Unsupervised Learning of Geometry with Edge-aware Depth-Normal Consistency [article]

Zhenheng Yang, Peng Wang, Wei Xu, Liang Zhao, Ramakant Nevatia
2017 arXiv   pre-print
Both layers are computed with awareness of edges inside the image to help address the issue of depth/normal discontinuity and preserve sharp edges.  ...  Specifically, we formulate an edge-aware depth-normal consistency term, and solve it by constructing a depth-to-normal layer and a normal-to-depth layer inside of the DCN.  ...  Conclusion In this paper, we propose an unsupervised learning framework for both depth and normal estimation via edge-aware depthnormal consistency.  ... 
arXiv:1711.03665v1 fatcat:wqtdxxk3dzdslej5t7da324hya

Unsupervised Learning Optical Flow by Robust Reconstruction and Edge-Aware Smoothing

Lingtong Kong, Jie Yang
2019 Australian Journal of Intelligent Information Processing Systems  
In this work, we present a novel unsupervised method that can learn optical flow by robust image reconstruction based on brightness constancy and structure similarity constraints.  ...  To overcome negative effects of occlusion and obtain natural dense pixel flow fields, we propose an edge-aware loss function for smoothing flow fields while following original image structure.  ...  Conclusion In this paper, we propose an unsupervised learning framework of optical flow which is robust in real scenes.  ... 
dblp:journals/ajiips/KongY19 fatcat:jjevcywajbdajcpijchvims3hm

Unsupervised Learning of Geometry From Videos With Edge-Aware Depth-Normal Consistency

Zhenheng Yang, Peng Wang, Wei Xu, Liang Zhao, Ramakant Nevatia
2018 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Both layers are computed with awareness of edges inside the image to help address the issue of depth/normal discontinuity and preserve sharp edges.  ...  In this paper, we propose to use surface normal representation for unsupervised depth estimation framework.  ...  Thus, we incorporate an edge-aware depth-normal consistency constraint inside the network which better regularizes the learning of depths (Sec. 4).  ... 
doi:10.1609/aaai.v32i1.12257 fatcat:23eodozk7zhxpm3tcchqjqtk24

Unsupervised Learning of Discriminative Edge Measures for Vehicle Matching between Nonoverlapping Cameras

Ying Shan, Harpreet S. Sawhney, Rakesh Kumar
2008 IEEE Transactions on Pattern Analysis and Machine Intelligence  
The weight of each match measure in the final decision is determined by a novel unsupervised learning process so that the same-different classification can be optimally separated in the combined measurement  ...  We employ a novel measurement vector consists of three independent edge-based measures and their associated robust measures computed from a pair of aligned vehicle edge maps.  ...  Any representations of civilians or civilian vehicles are an artifact of the experimental process and are used purely to expedite the research.  ... 
doi:10.1109/tpami.2007.70728 pmid:18276974 fatcat:nxwq5cvzynachhqtzgjsq7f3ha

Minority Resampling Boosted Unsupervised Learning with Hyperdimensional Computing for Threat Detection at the Edge of Internet of Things

Vivek Christopher, Tharmasanthiran Aathman, Kayathiri Mahendrakumaran, Rashmika Nawaratne, Daswin De Silva, Vishaka Nanayakkara, Damminda Alahakoon
2021 IEEE Access  
Next, the adaptation of the GSOM algorithm based on HD computing performs unsupervised learning from unlabeled data, within the bounds of the computational constraints of the Edge layer.  ...  More specifically, unsupervised machine learning methods are technically suited for the detection of behaviour-based cyber threat and attacks on IoT Edge as it can learn from unlabeled data [20] .  ...  She is presently the Director of the Centre for Open and Distance Learning at the University of Moratuwa.  ... 
doi:10.1109/access.2021.3111053 fatcat:ewqph2gvtfhijfqaafhnbmqk5m

Unsupervised Learning of Discriminative Edge Measures for Vehicle Matching between Non-Overlapping Cameras

Ying Shan, H.S. Sawhney, R.T. Kumar
2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05)  
The weight of each match measure in the final decision is determined by a novel unsupervised learning process so that the same-different classification can be optimally separated in the combined measurement  ...  We employ a novel measurement vector consists of three independent edge-based measures and their associated robust measures computed from a pair of aligned vehicle edge maps.  ...  Any representations of civilians or civilian vehicles are an artifact of the experimental process and are used purely to expedite the research.  ... 
doi:10.1109/cvpr.2005.358 dblp:conf/cvpr/ShanSK05 fatcat:ilteybaepvcntdtiyaneisifvq

R-VGAE: Relational-variational Graph Autoencoder for Unsupervised Prerequisite Chain Learning [article]

Irene Li, Alexander Fabbri, Swapnil Hingmire, Dragomir Radev
2020 arXiv   pre-print
Our method is notably the first graph-based model that attempts to make use of deep learning representations for the task of unsupervised prerequisite learning.  ...  The task of concept prerequisite chain learning is to automatically determine the existence of prerequisite relationships among concept pairs.  ...  Notably, it is the first graph-based model that attempts to make use of deep learning representations for the task of unsupervised prerequisite learning.  ... 
arXiv:2004.10610v1 fatcat:vefwnxirpfgnnldmwvxjoffcna

Unsupervised Training for Neural TSP Solver [article]

Elīza Gaile, Andis Draguns, Emīls Ozoliņš, Kārlis Freivalds
2022 arXiv   pre-print
To avoid these problems, we introduce a novel unsupervised learning approach.  ...  Our approach has the advantage over supervised learning of not requiring large labelled datasets.  ...  Unsupervised learning has the advantage over supervised learning of not needing large correctly labeled datasets.  ... 
arXiv:2207.13667v1 fatcat:gvk4orinlffrfmym2lfj3cwpzm

Improved Extraction of Objects from Urine Microscopy Images with Unsupervised Thresholding and Supervised U-net Techniques

Abdul Aziz, Harshit Pande, Bharath Cheluvaraju, Tathagato Rai Dastidar
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
To the best of our knowledge this is the first application of Deep Learning for extraction of clinically significant objects in urine microscopy images.  ...  Overall the proposed unsupervised method along with edge thresholding worked the best by extracting maximum number of objects and minimum number of artifacts.  ...  The higher performance of the proposed unsupervised method as compared to the Deep Learning-based U-net method is possibly due to lack of sufficiently large data for training the Deep Learning method.  ... 
doi:10.1109/cvprw.2018.00299 dblp:conf/cvpr/AzizPCD18 fatcat:4mwly43knnhr5al4pprfq2x5c4

PTE

Jian Tang, Meng Qu, Qiaozhu Mei
2015 Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '15  
One possible reason is that these text embedding methods learn the representation of text in a fully unsupervised way, without leveraging the labeled information available for the task.  ...  Predictive text embedding utilizes both labeled and unlabeled data to learn the embedding of text.  ...  Because of the unsupervised learning process, the representations learned through these text embedding models are general enough and can be applied to a variety of tasks such as classification, clustering  ... 
doi:10.1145/2783258.2783307 dblp:conf/kdd/TangQM15 fatcat:lmjjjaflz5cxflatwplflftvtq

Study on a New Method of Link-Based Link Prediction in the Context of Big Data

Chen Jicheng, Chen Hongchang, Li Hanchao, Fahd Abd Algalil
2021 Applied Bionics and Biomechanics  
A barrage of literature has been written to approach this problem; however, they mostly come from the angle of unsupervised learning (UL).  ...  This research is aimed at finding the most appropriate link-based link prediction methods in the context of big data based on supervised learning.  ...  Conflicts of Interest The authors declare no conflict of interest in the authorship of this article.  ... 
doi:10.1155/2021/1654134 fatcat:lzra62guyfdf7nkxkcfbqvujbi

Unsupervised Learning of Discriminative Attributes and Visual Representations

Chen Huang, Chen Change Loy, Xiaoou Tang
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
While most attribute learning methods are supervised by costly human-generated labels, we introduce a simple yet powerful unsupervised approach to learn and predict visual attributes directly from data  ...  The learned attributes are shown to be capable of encoding rich imagery properties from both natural images and contour patches.  ...  We employ a target of K +1 labels, all normalized to {0, 1} and consisting of one binary edge label and the K-bit attributes previously learned from the corresponding binary edge patch.  ... 
doi:10.1109/cvpr.2016.559 dblp:conf/cvpr/HuangLT16 fatcat:zt4qarlwobbslb52b5ylylelvu

Semi-Supervised Learning for Multi-Task Scene Understanding by Neural Graph Consensus [article]

Marius Leordeanu, Mihai Pirvu, Dragos Costea, Alina Marcu, Emil Slusanschi, Rahul Sukthankar
2020 arXiv   pre-print
The unsupervised learning process is repeated over several generations, in which each edge becomes a "student" and also part of different ensemble "teachers" for training other students.  ...  We address the challenging problem of semi-supervised learning in the context of multiple visual interpretations of the world by finding consensus in a graph of neural networks.  ...  Unsupervised learning over multiple iterations: The unsupervised learning experiments follow the steps of Algorithm 1: during the supervised stage (Step 1), we use the 8k labeled images (train set) to  ... 
arXiv:2010.01086v2 fatcat:ng27p5utdnabplogh5qhokdlh4
« Previous Showing results 1 — 15 out of 67,026 results