Structure-sensitive manifold ranking for video concept detection
Published 29 September 2007
Jinhui Tang, Xian‐Sheng Hua, Guo-Jun Qi, Meng Wang, Tao Mei, Xiuqing Wu
Citations49
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
This paper proposes a novel graph-based learning approach, named Structure-Sensitive Manifold Ranking (SSMR), based on a structure-sensitive similarity measure, which takes local distribution differences into account to more accurately measure pairwise similarity.
Abstract
Pairwise similarity of samples is an essential factor in graph propagation based semi-supervised learning methods. Usually it is estimated based on Euclidean distance. However, the structural assumption, which is a basic assumption in these methods, has not been taken into consideration in the normal pairwise similarity measure. In this paper, we propose a novel graph-based learning approach, named Structure-Sensitive Manifold Ranking (SSMR),based on a structure-sensitive similarity measure. Instead of using distance only, SSMR takes local distribution differences into account to more accurately measure pairwise similarity. Furthermore, we show that SSMR can also be deduced from a partial differential equation based anisotropic diffusion. Experiments conducted on the TRECVID dataset show that this approach significantly outperforms existing graph-based semi-supervised learning methods for video semantic concept detection.
Keywords
Computer Science
Matrix Analysis
22,234 Citations1985Roger A. Horn, Charles R. Johnson
IEEE Transactions on Pattern Analysis and Machine IntelligenceScale-space and edge detection using anisotropic diffusion
12,040 Citations1990Pietro Perona, Jitendra Malik
Combining labeled and unlabeled data with co-training
5,604 Citations1998Avrim Blum, Tom M. Mitchell
Advanced textbooks in control and signal processingPattern Classification
5,015 Citations2001Leo H. Chiang, Evan L. Russell +1 more
The MIT Press eBooksSemi-Supervised Learning
4,308 Citations2006Olivier Chapelle, Schölkopf, B. +1 more
This first comprehensive overview of semi-supervised learning presents state-of-the-art algorithms, a taxonomy of the field, selected applications, benchmark experiments, and perspectives on ongoing and future research.
MPG.PuRe (Max Planck Society)Learning with Local and Global Consistency
3,750 Citations2003Dengyong Zhou, Olivier Bousquet +3 more
A principled approach to semi-supervised learning is to design a classifying function which is sufficiently smooth with respect to the intrinsic structure collectively revealed by known labeled and unlabeled points.
Semi-supervised learning using Gaussian fields and harmonic functions
3,377 Citations2003Xiaojin Zhu, Zoubin Ghahramani +1 more
Manifold Regularization: A Geometric Framework for Learning from Labeled and Unlabeled Examples
3,266 Citations2006Mikhail Belkin, Partha Niyogi +1 more
A semi-supervised framework that incorporates labeled and unlabeled data in a general-purpose learner is proposed and properties of reproducing kernel Hilbert spaces are used to prove new Representer theorems that provide theoretical basis for the algorithms.
Pattern Classification
2,474 Citations2001Duda, Richard O, Hart, Peter E +1 more
Research Showcase @ Carnegie Mellon University (Carnegie Mellon University)Learning from Labeled and Unlabeled Data using Graph Mincuts
947 Citations2018Avrim Blum, Shuchi Chawla
An algorithm based on finding minimum cuts in graphs, that uses pairwise relationships among the examples in order to learn from both labeled and unlabeled data is considered.
Cambridge University Press eBooksGeometric Partial Differential Equations and Image Analysis
846 Citations2001Guillermo Sapiro
This paper presents a probabilistic procedure called "diffusion on non-flat manifolds" which automates the very labor-intensive and therefore time-heavy and expensive process of approximating the curvature of a surface.
Multiple Bernoulli relevance models for image and video annotation
813 Citations2004Siwei Feng, R. Manmatha +1 more
This work shows how it can do both automatic image annotation and retrieval (using one word queries) from images and videos using a multiple Bernoulli relevance model, which significantly outperforms previously reported results on the task of image and video annotation.
Semi-Supervised Self-Training of Object Detection Models
791 Citations2005Charles Rosenberg, Martial Hebert +1 more
The key contributions of this empirical study are to demonstrate that a model trained in this manner can achieve results comparable to a modeltrained in the traditional manner using a much larger set of fully labeled data, and that a training data selection metric that is defined independently of the detector greatly outperforms a selection metric based on the detection confidence generated by the detector.
Semi-supervised learning with graphs
649 Citations2005Xiaojin Zhu, John Lafferty +1 more
A series of novel semi-supervised learning approaches arising from a graph representation, where labeled and unlabeled instances are represented as vertices, and edges encode the similarity between instances are presented.
Encyclopedia of Machine Learning and Data MiningLearning from Labeled and Unlabeled Data
469 Citations2017Claude Sammut, Geoffrey I. Webb
Manifold-ranking based image retrieval
358 Citations2004Jingrui He, Mingjing Li +3 more
MRBIR first makes use of a manifold ranking algorithm to explore the relationship among all the data points in the feature space, and then measures relevance between the query and all the images in the database accordingly, which is different from traditional similarity metrics based on pair-wise distance.
Image annotation refinement using random walk with restarts
191 Citations2006Changhu Wang, Jing Feng +2 more
A novel approach to automatically refine the original annotations of images by using Random Walk with Restarts (RWR) to leverage both the corpus information and the original confidence information of the annotations.
MISSL
126 Citations2006Rouhollah Rahmani, Sally A. Goldman
This work presents MISSL (Multiple-Instance Semi-Supervised Learning) that transforms any MI problem into an input for a graph-based single-instance semi-supervised learning method that encodes the MI aspects of the problem simultaneously working at both the bag and point levels.
IEEE Transactions on Image ProcessingGeneralized Manifold-Ranking-Based Image Retrieval
118 Citations2006Jianhua He, Menghan Li +3 more
A general transductive learning framework named generalized manifold-ranking-based image retrieval (gMRBIR) for image retrieval that makes use of relevance feedback and active learning to refine the retrieval result so that it converges to the query concept as fast as possible.
Measure Based Regularization
118 Citations2003Olivier Bousquet, Olivier Chapelle +1 more
This paper proposes three theoretical methods for taking into account this distribution P(x) for regularization and provides links to existing graph-based semi-supervised learning algorithms.
Graph based multi-modality learning
104 Citations2005Hanghang Tong, Jingrui He +3 more
This paper reveals the regular optimization nature, transductive learning nature as well as prior fusion nature of the proposed schemes, respectively, and can be easily extended to unsupervised learning, including clustering and embedding.
Hidden Markov models for automatic annotation and content-based retrieval of images and video
85 Citations2005Arnab Ghoshal, Pavel Ircing +1 more
A novel method for automatic annotation of images with keywords from a generic vocabulary of concepts or objects for the purpose of content-based image retrieval and results are presented on two image-collections | COREL and key-frames from TRECVID.
Statistical models for automatic video annotation and retrieval
75 Citations2004Victor Lavrenko, Siwei Feng +1 more
A continuous relevance model is applied to the problem of directly retrieving the visual content of videos using text queries and a modified model is proposed - the normalized CRM - which substantially improves performance on a subset of the TREC video dataset.
Semi-Supervised Cross Feature Learning for Semantic Concept Detection in Videos
75 Citations2005Rong Yan, Milind Naphade
The proposed SCFL algorithm not only significantly outperforms the conventional co-training algorithms, but also comes close to achieving the performance when the unlabeled set were to be manually annotated and used for training along with the labeled data set.
Semi-automatic video annotation based on active learning with multiple complementary predictors
51 Citations2005Yan Song, Xian‐Sheng Hua +2 more
It is proved that the samples selected by the proposed scheme are more representative than general active learning scheme, as well as the incremental model adaptation scheme is effective especially when the newly added data size is small.
Automatic video annotation by semi-supervised learning with kernel density estimation
45 Citations2006Meng Wang, Xian‐Sheng Hua +4 more
This paper proposes a novel semi-supervised learning algorithm, named Semi Supervised Learning by Kernel Density Estimation (SSLKDE), which is based on a non-parametric method, and therefore the "model assumption" is avoided.
Manifold-ranking based video concept detection on large database and feature pool
44 Citations2006Xun Yuan, Xian‐Sheng Hua +2 more
A novel manifold- ranking based scheme is proposed, which consists of three major components: feature pool construction, pre-filtering, and manifold-ranking, which is extensible and flexible in terms of adding new features into the feature pool, introducing human interactions on selecting features, and defining new concepts.
Electronics LettersVideo annotation based on temporally consistent Gaussian random field
24 Citations2007Jinhui Tang, Xian‐Sheng Hua +3 more
A novel method for automatically annotating video semantics, called temporally consistent Gaussian random field (TCGRF), which adapts the temporal consistency property of video data into graph-based semi-supervised learning to improve the annotation results.
Video annotation by graph-based learning with neighborhood similarity
23 Citations2007Meng Wang, Tao Mei +3 more
A novel neighborhood similarity measure, which simultaneously takes into account both the distance between samples and the difference between the structures around the corresponding samples, which demonstrates that the neighborhood similarity is superior to existing distance based similarity.
Anisotropic Manifold Ranking for Video Annotation
8 Citations2007Jinhui Tang, Xian‐Sheng Hua +3 more
This work proposes a novel graph-based SSL method for video annotation, named anisotropic manifold ranking (AniMR), based on a structure-related similarity measure, and demonstrates that the label propagation in AniMR is aniMR, which is intrinsically different from the isotropic label propagation process in general graph- based SSL methods.