Filters








35,541 Hits in 3.5 sec

Classifier based graph construction for video segmentation

Anna Khoreva, Fabio Galasso, Matthias Hein, Bernt Schiele
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Graph-based methods, enabling topperformance on recent benchmarks, consist of three essential components: 1. powerful features account for object appearance and motion similarities; 2. spatio-temporal  ...  While a wide variety of features have been explored and various graph partition algorithms have been proposed, there is surprisingly little research on how to construct a graph to obtain the best video  ...  To the best of our knowledge, graph construction based on classifier-learnt combination of features is novel in video segmentation.  ... 
doi:10.1109/cvpr.2015.7298697 dblp:conf/cvpr/KhorevaG0S15 fatcat:i3mekzwiubasrmobrejh4l7oba

ACGNet: Action Complement Graph Network for Weakly-Supervised Temporal Action Localization

Zichen Yang, Jie Qin, Di Huang
2022 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
In this paper, we tackle this problem from a new perspective by enhancing segment-level representations with a simple yet effective graph convolutional network, namely action complement graph network (  ...  It facilitates the current video segment to perceive spatial-temporal dependencies from others that potentially convey complementary clues, implicitly mitigating the negative effects caused by the two  ...  The action complement graph (ACG) is constructed for each video in a principled way to exchange complementary information between its nodes (i.e., segments).  ... 
doi:10.1609/aaai.v36i3.20216 fatcat:brxtspeelrcbjnyselb3dj5oei

A Study on Relational Semantic Video Content Extraction Using Bayesian Network Classifier

S Rajarajeshwari, G.Kalaimathi Priya, S.Grace Mary, Dr.V.Sambath kumar
2014 IOSR Journal of Computer Engineering  
The VISCOM is used to construct ontology for a given domain and the rule based model is used to define some complex situation more effectively.  ...  The content based extraction in videos is an important application due to the rapid growth in the video based application.  ...  If the resulted segmented video is over segmented the subsequent region merging operation is done and the region adjacency graph is constructed.  ... 
doi:10.9790/0661-1631106109 fatcat:5zt3umncmba4tft3dob7siqjga

Ensemble Video Object Cut in Highly Dynamic Scenes

Xiaobo Ren, Tony X. Han, Zhihai He
2013 2013 IEEE Conference on Computer Vision and Pattern Recognition  
We incorporate this similarity information into a graph-cut energy minimization framework for foreground object segmentation.  ...  We consider video object cut as an ensemble of framelevel background-foreground object classifiers which fuses information across frames and refine their segmentation results in a collaborative and iterative  ...  Foreground Salience Graph. Based on the temporal and spatial salience measures, we can then construct the foreground salience graph.  ... 
doi:10.1109/cvpr.2013.254 dblp:conf/cvpr/RenHH13 fatcat:va5zydzufbe33ixhfajpekrkzu

Indian classical dance action identification using adaptive graph matching from unconstrained videos

K V.V. Kumar, P V.V. Kishore
2017 International Journal of Engineering & Technology  
A new segmentation model is developed using discrete wavelet transform and local binary pattern features for segmentation.  ...  The extracted features were represented as a graph and a novel adaptive graph matching algorithm is proposed.  ...  For recognition, an adaptive graph matching algorithm is proposed to classify query dance video based on the dance dataset.  ... 
doi:10.14419/ijet.v7i1.1.10156 fatcat:jsuzbgkwhbfevcm4ke57rrzu6y

ACGNet: Action Complement Graph Network for Weakly-supervised Temporal Action Localization [article]

Zichen Yang, Jie Qin, Di Huang
2021 arXiv   pre-print
In this paper, we tackle this problem from a new perspective by enhancing segment-level representations with a simple yet effective graph convolutional network, namely action complement graph network (  ...  It facilitates the current video segment to perceive spatial-temporal dependencies from others that potentially convey complementary clues, implicitly mitigating the negative effects caused by the two  ...  The action complement graph interference provide relatively stable information and low- (ACG) is constructed for each video in a principled way quality segments can also be complementary  ... 
arXiv:2112.10977v1 fatcat:r5xj6pnstzelzbhysluzscf5vq

Segmentation Improved Label Propagation for Semi-supervised Anomaly Detection in Fused Magnesia Furnace Process

Shaowen Lu
2020 IEEE Access  
The second part is to train a LSTM classifier for the detection of semi-molten working condition. The classifier is trained under the framework of graph based label propagation.  ...  The graph based label propagation is used to iterative spread the labeling information and train a LSTM classifier.  ... 
doi:10.1109/access.2020.3042464 fatcat:54x5owbicba2netihg7kx4ns3q

Action recognition in video using a spatial-temporal graph-based feature representation

Iveel Jargalsaikhan, Suzanne Little, Remi Trichet, Noel E. O'Connor
2015 2015 12th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)  
We propose a video graph based human action recognition framework.  ...  In particular, we extend a popular dbscan density-based clustering algorithm to form an intuitive video graph.  ...  Graph-cut based methods have achieved impressive performance for object segmentation, even on difficult image datasets [6] .  ... 
doi:10.1109/avss.2015.7301760 dblp:conf/avss/JargalsaikhanLT15 fatcat:o3ni4voco5eq5lucohurfmorra

A Generic Framework for Video Annotation via Semi-Supervised Learning

Tianzhu Zhang, Changsheng Xu, Guangyu Zhu, Si Liu, Hanqing Lu
2012 IEEE transactions on multimedia  
Concretely, a Fast Graph-based Semi-Supervised Multiple Instance Learning (FGSSMIL) algorithm, which aims to simultaneously tackle these difficulties in a generic framework for various video domains (e.g  ...  Two critical issues of FGSSMIL are: 1) how to calculate the weight assignment for a graph construction, where the weight of an edge specifies the similarity between two data points.  ...  Similarity Measure for Graph Construction More recently, graph-based methods have attracted the interest of researchers in this community due to their effectiveness and computational efficiency (most graph-based  ... 
doi:10.1109/tmm.2012.2191944 fatcat:7uaujwzq4nfrto5jaim7bf4ify

Self-propagating video segmentation using patch matching and enhanced Onecut

Shoudong Han, Chenfei Xia, Yujun Liu, Yingchun Yang
2020 SN Applied Sciences  
constructed to obtain its segmentation result.  ...  In this paper, a self-propagating video segmentation approach based on patch matching and enhanced Onecut is proposed, which takes full advantage of the target's color feature, shape feature, and motion  ...  Enhanced Onecut segmentation model In recent years, video segmentation method based on graph cuts framework [24] [25] [26] [27] has become a research hotspot.  ... 
doi:10.1007/s42452-020-2033-8 fatcat:pi6z32f2gndhjab2gbvprueyzq

Graph-based foreground extraction in extended color space

Hansung Kim, Adrian Hilton
2009 2009 16th IEEE International Conference on Image Processing (ICIP)  
For efficient computation, the graph structure is optimized by the minimum spanning tree before segmentation.  ...  Then the foreground region is extracted with a graph-based region segmentation method considering background difference and spatial homogeneity.  ...  Mu et al. have proposed a combination method of block-based initial segmentation and refinement by Graph-cut for automatic segmentation [7] .  ... 
doi:10.1109/icip.2009.5414370 dblp:conf/icip/KimH09 fatcat:iwb6cy2q5zemjorwhpgq25iqqa

FaceSeg: Automatic Face Segmentation for Real-Time Video

Hongliang Li, K.N. Ngan, Qiang Liu
2009 IEEE transactions on multimedia  
In this paper, we present an accurate segmentation system for cutting human faces out from video sequences in real-time.  ...  Segmenting human faces automatically is very important for face recognition and verification, security system, and computer vision.  ...  A statistical model-based video segmentation for head-andshoulder type video is addressed in [16] .  ... 
doi:10.1109/tmm.2008.2008922 fatcat:ozoih5kj2rfr7lzbdy4huqwtkm

Fuzzy Ontology and Rule based Model for Automatic Semantic Content Extraction from Videos using k-Means Algorithm

Priyanka Nikam, B.R. Nandwalkar
2015 International Journal of Computer Applications  
Video based applications disclosed need for efficiently extracting and modeling the video contents. The video features can be classified into normal data, relative features and logic content.  ...  Semantic level understanding is required for core content of video. So to get video content automatic semantic content framework is proposed.  ...  Fig3 shows resuts for frame segmention of a given video. i.e. number of segments done in a frame. In proposed system frame is divided into less number of segments than genetic based approach.  ... 
doi:10.5120/ijca2015907148 fatcat:uckz54b5mbgapfzj3bn6emcbbq

A generic framework for event detection in various video domains

Tianzhu Zhang, Changsheng Xu, Guangyu Zhu, Si Liu, Hanqing Lu
2010 Proceedings of the international conference on Multimedia - MM '10  
Concretely, a Graph-based Semi-Supervised Multiple Instance Learning (GSSMIL) algorithm is proposed to jointly explore small-scale expert labeled videos and large-scale unlabeled videos to train the event  ...  A critical issue of GSSMIL in constructing a graph is the weight assignment, where the weight of an edge specifies the similarity between two data points.  ...  To effectively obtain the similarity measure for the graph construction, we present an MILIS measure by considering the class structure.  ... 
doi:10.1145/1873951.1873967 dblp:conf/mm/ZhangXZLL10 fatcat:aznsm4tedzak3nn4ifq22x2p3y

Improved Image Boundaries for Better Video Segmentation [chapter]

Anna Khoreva, Rodrigo Benenson, Fabio Galasso, Matthias Hein, Bernt Schiele
2016 Lecture Notes in Computer Science  
Graph-based video segmentation methods rely on superpixels as starting point.  ...  While most previous work has focused on the construction of the graph edges and weights as well as solving the graph partitioning problem, this paper focuses on better superpixels for video segmentation  ...  As our approach is directly applicable to any graph-based video segmentation technique we additionaly evaluated our superpixels with the classifier-based graph construction method of [24] .  ... 
doi:10.1007/978-3-319-49409-8_64 fatcat:ow7hvcxcvva7lj5v24ioa25rli
« Previous Showing results 1 — 15 out of 35,541 results