Filters








182,972 Hits in 6.2 sec

Learning Multiple Visual Tasks while Discovering their Structure [article]

Carlo Ciliberto, Lorenzo Rosasco, Silvia Villa
2015 arXiv   pre-print
The key idea is that exploring task relatedness (structure) can lead to improved performances.  ...  of multiple agents, or denoising, to name a few.  ...  Conclusions We proposed a learning framework designed to solve multiple related tasks while simultaneously recovering their structure.  ... 
arXiv:1504.03106v1 fatcat:q6yprek425gz5hgfat6zbl23wu

Learning multiple visual tasks while discovering their structure

Carlo Ciliberto, Lorenzo Rosasco, Silvia Villa
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
The key idea is that exploring task relatedness (structure) can lead to improved performances.  ...  of multiple agents, or denoising, to name a few.  ...  We proposed a learning framework designed to solve multiple related tasks while simultaneously recovering their structure.  ... 
doi:10.1109/cvpr.2015.7298608 dblp:conf/cvpr/CilibertoRV15 fatcat:7fbrulnmibgopovzx76ocb23w4

Visual Understanding of Multiple Attributes Learning Model of X-Ray Scattering Images [article]

Xinyi Huang, Suphanut Jamonnak, Ye Zhao, Boyu Wang, Minh Hoai, Kevin Yager, Wei Xu
2019 arXiv   pre-print
This extended abstract presents a visualization system, which is designed for domain scientists to visually understand their deep learning model of extracting multiple attributes in x-ray scattering images  ...  The system focuses on studying the model behaviors related to multiple structural attributes.  ...  The same structural attributes (patterns) can be of great variety in their appearances.  ... 
arXiv:1910.04357v1 fatcat:3zxpwwd3azddbl4r2talylxr7i

Visualizing Semantic Structures of Sequential Data by Learning Temporal Dependencies [article]

Kyoung-Woon On, Eun-Sol Kim, Yu-Jung Heo, Byoung-Tak Zhang
2019 arXiv   pre-print
The temporal dependency structure of semantic is discovered by learning parameterized kernels of graph convolutional methods.  ...  In addition, the semantic structures within given sequential data can be interpreted by visualizing temporal dependencies learned from the method.  ...  In this work, we propose a graph-based neural network architecture, which learns representations of sequential data while considering the temporal dependency structure of it.  ... 
arXiv:1901.09066v1 fatcat:csumoltherfvlmtdws63vdamim

Unified Perceptual Parsing for Scene Understanding [article]

Tete Xiao, Yingcheng Liu, Bolei Zhou, Yuning Jiang, Jian Sun
2018 arXiv   pre-print
Humans recognize the visual world at multiple levels: we effortlessly categorize scenes and detect objects inside, while also identifying the textures and surfaces of the objects along with their different  ...  A multi-task framework called UPerNet and a training strategy are developed to learn from heterogeneous image annotations.  ...  called UPerNet with hierarchical structure to learn from heterogeneous data from multiple image datasets. 3) The model is shown to be able to jointly infer and discover the rich visual knowledge underneath  ... 
arXiv:1807.10221v1 fatcat:knzcor2b3vcb3gn6clp5yh3q3a

Interactive Visualization of AI-based Speech Recognition Texts

Tsung Heng Wu, Ye Zhao, Md Amiruzzaman
2020 International EuroVis Workshop on Visual Analytics  
In this paper, we show interactive visualization can play important roles in post-AI understanding, editing, and analysis of speech recognition results by presenting specified task characterization and  ...  Speech recognition technology has achieved impressive success recently with AI techniques of deep learning networks.  ...  While most techniques work on vision datasets with a focus on CNNs (convolutional neural networks), RNN visualization tools are developed for linguistic, biological, and vision tasks.  ... 
doi:10.2312/eurova.20201091 dblp:conf/eurova-ws/WuZA20 fatcat:6n3iuuf7xjc3tb7faqwt3h5wli

Minimally invasive surgery training using multiple port sites to improve performance

Alan D. White, Oscar Giles, Rebekah J. Sutherland, Oliver Ziff, Mark Mon-Williams, Richard M. Wilkie, J. Peter A. Lodge
2013 Surgical Endoscopy  
Structural learning theory suggests that experiencing motor task variation enables the central nervous system to extract general rules regarding tasks with a similar structurerules which subsequently can  ...  The purpose of the present study was to determine if structural learning theory can be applied to MIS to inform training methods.  ...  Once the driver has discovered the structure, the problem of learning a related task (driving a new car) becomes much simpler [7] .  ... 
doi:10.1007/s00464-013-3307-7 pmid:24232133 fatcat:ngbzwcjsobc7lcml35yczqylgi

Salient region detection and segmentation for general object recognition and image understanding

TieJun Huang, YongHong Tian, Jia Li, HaoNan Yu
2011 Science China Information Sciences  
By exploiting multi-task learning methods to model visual saliency simultaneously with the bottom-up and top-down factors, the lowest layer can effectively detect salient objects in an image.  ...  ., salient region detection, object segmentation, automatic object discovering and visual dictionary construction.  ...  Learning-based salient region detection Typically, the visual world is highly structured in a 3D manner.  ... 
doi:10.1007/s11432-011-4487-1 fatcat:nmbyczekrjdwbpnjzix2v4p4ti

Unified Perceptual Parsing for Scene Understanding [chapter]

Tete Xiao, Yingcheng Liu, Bolei Zhou, Yuning Jiang, Jian Sun
2018 Lecture Notes in Computer Science  
Humans recognize the visual world at multiple levels: we effortlessly categorize scenes and detect objects inside, while also identifying the textures and surfaces of the objects along with their different  ...  The trained networks are further applied to discover visual knowledge in natural scenes 1 .  ...  called UPerNet with hierarchical structure to learn from heterogeneous data from multiple image datasets. 3) The model is shown to be able to jointly infer and discover the rich visual knowledge underneath  ... 
doi:10.1007/978-3-030-01228-1_26 fatcat:bopejhyckrfwpe7575xt2vz2je

Interactive Visual Study of Multiple Attributes Learning Model of X-Ray Scattering Images [article]

Xinyi Huang, Suphanut Jamonnak, Ye Zhao, Boyu Wang, Minh Hoai, Kevin Yager, Wei Xu
2020 arXiv   pre-print
However, visual analytics tools are lacking for the specific application of x-ray image classification with multiple structural attributes.  ...  In this paper, we present an interactive system for domain scientists to visually study the multiple attributes learning models applied to x-ray scattering images.  ...  Effective visualization tools are needed to discover the relationship among the scientific structural objects on scientific images and the performance of deep learning models.  ... 
arXiv:2009.02256v1 fatcat:hvwf6dumazaodca26dwdtcyate

Discovering Synergies for Robot Manipulation with Multi-Task Reinforcement Learning [article]

Zhanpeng He, Matei Ciocarlie
2022 arXiv   pre-print
We also show that deriving synergies using multiple tasks can lead to a subspace that enables robots to efficiently learn new manipulation tasks and interactions with new objects.  ...  In this paper, we present a framework that simultaneously discovers a synergy space and a multi-task policy that operates on this low-dimensional action space to accomplish diverse manipulation tasks.  ...  DISCUSSION 1) Can DiscoSyn discover synergies while simultaneously learning tasks?  ... 
arXiv:2110.01530v2 fatcat:vxwbbchvtbfzpmil5txs4jkmb4

Unsupervised Learning from Narrated Instruction Videos

Jean-Baptiste Alayrac, Piotr Bojanowski, Nishant Agrawal, Josef Sivic, Ivan Laptev, Simon Lacoste-Julien
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We address the problem of automatically learning the main steps to complete a certain task, such as changing a car tire, from a set of narrated instruction videos.  ...  Third, we experimentally demonstrate that the proposed method can automatically discover, in an unsupervised manner, the main steps to achieve the task and locate the steps in the input videos.  ...  In this work, we consider instruction videos and develop a method that learns a sequence of steps, as well as their textual and visual representations, required to achieve a certain task.  ... 
doi:10.1109/cvpr.2016.495 dblp:conf/cvpr/AlayracBASLL16 fatcat:gohmry3c2vhjjff2s7d3gofsxy

Data-Driven Activities Involving Electronic Health Records: An Activity and Task Analysis Framework for Interactive Visualization Tools

Neda Rostamzadeh, Sheikh S. Abdullah, Kamran Sedig
2020 Multimodal Technologies and Interaction  
The survey includes an overview of the goal of each IVT, a brief description of its visualization, and an analysis of how sub-activities, tasks, and sub-tasks blend and combine to accomplish the tool's  ...  In this paper, we present a framework to identify and analyze EHR-data-driven tasks and activities in the context of interactive visualization tools (IVTs)—that is, all the activities, sub-activities,  ...  Acknowledgments: We would like to thank all authors and publishers who shared images of their tools with us. Conflicts of Interest: The authors declare that there is no conflict of interest.  ... 
doi:10.3390/mti4010007 fatcat:qlnhoyt7bnga3bqhjn6ibtr33e

Learning of speech categories in humans and Zebra finches [article]

Dimitrios Botskaris, Pralle Kriengwatana, Carel ten Cate
2016 bioRxiv   pre-print
Twelve Greek listeners and four Zebra finches were tested in speech category learning tasks.  ...  If Zebra Finches are actually able to acquire (RB) and (II) category structures using the same strategies as humans, the utility of multiple systems of categorization might not be restricted to primates  ...  All species struggled to integrate information across two stimulus dimensions, while their accuracy improved significantly on tasks that required only one single dimension, showing that process of learning  ... 
doi:10.1101/077321 fatcat:ptvtztvpxfardfe4najbe5hdwu

Interactively building a discriminative vocabulary of nameable attributes

Devi Parikh, Kristen Grauman
2011 CVPR 2011  
We demonstrate the approach with multiple datasets, and show its clear advantages over baselines that lack a nameability model or rely on a list of expert-provided attributes.  ...  Human-nameable visual attributes offer many advantages when used as mid-level features for object recognition, but existing techniques to gather relevant attributes can be inefficient (costing substantial  ...  Hence, we propose to first discover visually discriminative features as potential attributes, and then determine their nameability.  ... 
doi:10.1109/cvpr.2011.5995451 dblp:conf/cvpr/ParikhG11 fatcat:nyxcokgayrbz7fhdihjqidhmtq
« Previous Showing results 1 — 15 out of 182,972 results