Filters








102 Hits in 2.5 sec

First Person Action-Object Detection with EgoNet [article]

Gedas Bertasius, Hyun Soo Park, Stella X. Yu, Jianbo Shi
2017 arXiv   pre-print
In this paper, we study the tight interplay between our momentary visual attention and motor action with objects from a first-person camera.  ...  Action-objects may be task-dependent but since many tasks share common person-object spatial configurations, action-objects exhibit a characteristic 3D spatial distance and orientation with respect to  ...  action-object detection from first-person RGBD data.  ... 
arXiv:1603.04908v3 fatcat:vllevk4xpfgzjnoxr625tqokzu

First-Person Action-Object Detection with EgoNet

Gedas Bertasius, Hyun Soo Park, Stella Yu, Jianbo Shi
2017 Robotics: Science and Systems XIII  
In this paper, we study the tight interplay between our momentary visual attention and motor action with objects from a first-person camera.  ...  Action-objects may be task-dependent but since many tasks share common person-object spatial configurations, action-objects exhibit a characteristic 3D spatial distance and orientation with respect to  ...  action-object detection from first-person RGBD data.  ... 
doi:10.15607/rss.2017.xiii.012 dblp:conf/rss/BertasiusPYS17 fatcat:cvgtb526drcq5bitbldxwbnmge

Unsupervised Learning of Important Objects from First-Person Videos [article]

Gedas Bertasius, Hyun Soo Park, Stella X. Yu, Jianbo Shi
2017 arXiv   pre-print
In this work, we show that we can detect important objects in first-person images without the supervision by the camera wearer or even third-person labelers.  ...  Most prior methods for this task learn to detect such important objects from the manually labeled first-person data in a supervised fashion.  ...  We believe that in the future, our method could be extended to other tasks such as first-person activity recognition, or egocentric video summarization.  ... 
arXiv:1611.05335v3 fatcat:lb6mxbq4nfcs3dlb4uigjwxhki

oddball: Spotting Anomalies in Weighted Graphs [chapter]

Leman Akoglu, Mary McGlohon, Christos Faloutsos
2010 Lecture Notes in Computer Science  
million nodes, where OddBall indeed spots unusual nodes that agree with intuition.  ...  following: (a) we discover several new rules (power laws) in density, weights, ranks and eigenvalues that seem to govern the socalled "neighborhood sub-graphs" and we show how to use these rules for anomaly detection  ...  This work is also partially supported by an IBM Faculty Award, a SPRINT gift, with additional funding from Intel, and Hewlett-Packard.  ... 
doi:10.1007/978-3-642-13672-6_40 fatcat:xym6jah6lvbhxmvjbfkpjk2674

Anomaly Detection in Graphs of Bank Transactions for Anti Money Laundering Applications

Bogdan Dumitrescu, Andra Baltoiu, Stefania Budulan
2022 IEEE Access  
Our method is based on designing new features; the most important are those resulting from the reduced egonet, which is the subgraph that remains from an egonet after eliminating the nodes connected with  ...  Our features are added to usual egonet features and a general anomaly detection algorithm, in our case Isolation Forest, serves to detect the anomalies.  ...  LEARNING APPROACHES More abstract approaches are based on learning using a global objective function. Our method has no resemblance with them. A first example is that of building embeddings.  ... 
doi:10.1109/access.2022.3170467 fatcat:vbblqcmyczgmfop2blz6gt5i7a

A Survey on Different Graph Based Anomaly Detection Techniques

Debajit Sensarma, Samar Sen Sarma
2015 Indian Journal of Science and Technology  
Finally, the relevance of cyber crime and its elimination is highlighted throughout the paper with some real world applications of graph based anomaly detection techniques and also some future direction  ...  to improve the technique of detecting anomalies in data has been given.  ...  It is basically based upon the theory that a person attempting to commit an unusual or illegal action would do so by imitating the known behaviors thus concealing their true intensions.  ... 
doi:10.17485/ijst/2015/v8i1/75197 fatcat:2eckrpzh6va7dmixqooukngtly

Ajalon: Simplifying the authoring of wearable cognitive assistants

Truong An Pham, Junjue Wang, Roger Iyengar, Yu Xiao, Padmanabhan Pillai, Roberta Klatzky, Mahadev Satyanarayanan
2021 Software, Practice & Experience  
It is inspired by, and broadens, the metaphor of GPS navigation tools that provide real-time step-by-step guidance, with prompt error detection and correction.  ...  Although the average savings factor for this step (1.43) is comparable to that of the data labeling component of building object detectors with OpenTPOD, the absolute time benefit of using Ajalon to extract  ...  On average, subjects spent 0.9 h without PTEditor while spending 0.4 h with it, a significant difference, (F 1, 6 = 64.374, p < 0.001).  ... 
doi:10.1002/spe.2987 fatcat:qrn2icargjakfasl5b5duk5dpy

Towards business partnership recommendation using user opinion on Facebook

Diego P. Tsutsumi, Amanda T. Fenerich, Thiago H. Silva
2019 Journal of Internet Services and Applications  
Besides, we propose an algorithm for detecting business communities in the considered model.  ...  We also propose an algorithm to identify possible business outliers in the detected communities, which could represent an automatic way to identify non-obvious relations that might deserve particular attention  ...  First, it is proposed in this study a new approach to extract relations outliers on the communities detected.  ... 
doi:10.1186/s13174-019-0110-2 fatcat:cgkrkywxhzfsllyug3zbwzly7a

Graph based anomaly detection and description: a survey

Leman Akoglu, Hanghang Tong, Danai Koutra
2014 Data mining and knowledge discovery  
As objects in graphs have long-range correlations, a suite of novel technology has been developed for anomaly detection in graph data.  ...  Detecting anomalies in data is a vital task, with numerous high-impact applications in areas such as security, finance, health care, and law enforcement.  ...  occur, based on their relationships with other objects.  ... 
doi:10.1007/s10618-014-0365-y fatcat:rfjn7bwdgra5faorwbdkkb45ze

Graph-based Anomaly Detection and Description: A Survey [article]

Leman Akoglu and Hanghang Tong and Danai Koutra
2014 arXiv   pre-print
As objects in graphs have long-range correlations, a suite of novel technology has been developed for anomaly detection in graph data.  ...  Detecting anomalies in data is a vital task, with numerous high-impact applications in areas such as security, finance, health care, and law enforcement.  ...  occur, based on their relationships with other objects.  ... 
arXiv:1404.4679v2 fatcat:y6nsswymcfc2pa7qe7zrjzc7wq

A Self Validation Network for Object-Level Human Attention Estimation [article]

Zehua Zhang, Chen Yu, David Crandall
2019 arXiv   pre-print
Estimating this object of attention in first-person (egocentric) videos is useful for many human-centered real-world applications such as augmented reality applications and driver assistance systems.  ...  are generated from an off-the-shelf object detector.  ...  EgoNet [5] , among the first papers to focus on important object detection in first-person videos, combines visual appearance and 3D layout information to generate probability maps of object importance  ... 
arXiv:1910.14260v2 fatcat:dr2xqgxy75cgngjzqkpheorcfa

PERSEUS-HUB: Interactive and Collective Exploration of Large-Scale Graphs

Di Jin, Aristotelis Leventidis, Haoming Shen, Ruowang Zhang, Junyue Wu, Danai Koutra
2017 Informatics  
that are worth investigating, and provides users with uncluttered visualization and easy interaction with complex graph statistics.  ...  patterns of interest with rich side information and discovering relations within the data.  ...  (kNN) technique to measure how isolated an object is with respect to its neighborhood.  ... 
doi:10.3390/informatics4030022 fatcat:7tlm5oajuvbdnh552atwy5bjpa

H+O: Unified Egocentric Recognition of 3D Hand-Object Poses and Interactions

Bugra Tekin, Federica Bogo, Marc Pollefeys
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Given a single RGB image, our model jointly estimates the 3D hand and object poses, models their interactions, and recognizes the object and action classes with a single feed-forward pass through a neural  ...  The complete model takes as input a sequence of frames and outputs per-frame 3D hand and object pose predictions along with the estimates of object and action categories for the entire sequence.  ...  EgoNet [2] detects "action objects", i.e. objects linked to visual or tactile interactions, from firstperson RGB-D images. Fathi et al.  ... 
doi:10.1109/cvpr.2019.00464 dblp:conf/cvpr/TekinBP19 fatcat:xz7iop75wvdybmqbtyaprvhwru

Network and actor attribute effects on the performance of researchers in two fields of social science in a small peripheral community

Srebrenka Letina
2016 Journal of Informetrics  
We used log-odds to demonstrate the probabilities of the outcome for three prototypical egonet structures: open, closed and complex; with different numbers of alters with attribute.  ...  Employing the auto-logistic actor attribute models allowed the inclusion of six actor attributes and the analysis of their effects simultaneously with network effects.  ...  Introduction Scientists are often perceived, personally and professionally, as "solitary minds", with high autonomy, freedom and independence (Fox & Faver, 1984 ).  ... 
doi:10.1016/j.joi.2016.03.007 fatcat:nosizrpx2fh3jbuucebifkcuxe

Is First Person Vision Challenging for Object Tracking? [article]

Matteo Dunnhofer, Antonino Furnari, Giovanni Maria Farinella, Christian Micheloni
2021 arXiv   pre-print
Understanding human-object interactions is fundamental in First Person Vision (FPV).  ...  In this paper, we fill the gap by presenting the first systematic study of object tracking in FPV.  ...  First-person action-object detection with egonet. In [17] Dorin Comaniciu, Visvanathan Ramesh, and Peter Meer. Proceedings of Robotics: Science and Systems, July 2017.  ... 
arXiv:2108.13665v1 fatcat:hexje3x6t5cdpbmj4l26m5h3lq
« Previous Showing results 1 — 15 out of 102 results