VisionGo: Towards video retrieval with joint exploration of human and computer

Luan, Huanbo; Yong, Zheng; Wang, Meng; Chua, Tat-Seng

doi:10.1016/j.ins.2011.05.018

Cited by 18 publications

(5 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The VisionGo system [Luan et al 2011] aims at maximizing the effectiveness of human annotators. An initial automatically ranked set of video shots, based on a textual query, can be refined by an interactive process.…”

Section: Keyboard and Mousementioning

confidence: 99%

Video Interaction Tools

Schoeffmann

Hudelist

Huber³

2015

ACM Comput. Surv.

View full text Add to dashboard Cite

Digital video enables manifold ways of multimedia content interaction. Over the last decade, many proposals for improving and enhancing video content interaction were published. More recent work particularly leverages on highly capable devices such as smartphones and tablets that embrace novel interaction paradigms, for example, touch, gesture-based or physical content interaction. In this article, we survey literature at the intersection of Human-Computer Interaction and Multimedia. We integrate literature from video browsing and navigation, direct video manipulation, video content visualization, as well as interactive video summarization and interactive video retrieval. We classify the reviewed works by the underlying interaction method and discuss the achieved improvements so far. We also depict a set of open problems that the video interaction community should address in future.

show abstract

Section: Keyboard and Mousementioning

confidence: 99%

Video Interaction Tools

Schoeffmann

Hudelist

Huber³

2015

ACM Comput. Surv.

View full text Add to dashboard Cite

show abstract

“…However, it is well known that points, which are close to each other in one context, may appear quite distant in another context. This is why modification of similarity measure is a fairly common way to adapt to users and contexts [ 62 , 63 ]. Furthermore, experimental comparison of conventional semisupervised learning with more lightweight context adaptation demonstrated that the latter can be significantly more accurate [ 37 ].…”

Section: Basic Approaches and Examples Of The Lightweight Adaptatimentioning

confidence: 99%

“…(3) Tuning classifiers for small datasets : special efforts are taken for selecting data features, classifier parameters, or training samples to reduce negative effect of small data size. This approach was proposed for multimedia retrieval with SVM (support vector machine), and tuning was performed by selecting SVM kernel or subset of training items [ 63 , 73 ]. SVM training was done for each query in a standard way, by minimising the classification error for the user-labelled items.…”

Section: Basic Approaches and Examples Of The Lightweight Adaptatimentioning

confidence: 99%

“…In particular, an iterative user's feedback in [ 73 ] required every user to label nine retrieved images per iteration and achieved a reasonable precision after 6–8 iterations. Video retrieval in [ 63 ] gave the users 15 minutes per query for evaluating and labelling the obtained query results.…”

Section: Basic Approaches and Examples Of The Lightweight Adaptatimentioning

confidence: 99%

See 1 more Smart Citation

Lightweight Adaptation of Classifiers to Users and Contexts: Trends of the Emerging Domain

Vildjiounaite

Gimel’farb

Kyllönen

et al. 2015

The Scientific World Journal

View full text Add to dashboard Cite

Intelligent computer applications need to adapt their behaviour to contexts and users, but conventional classifier adaptation methods require long data collection and/or training times. Therefore classifier adaptation is often performed as follows: at design time application developers define typical usage contexts and provide reasoning models for each of these contexts, and then at runtime an appropriate model is selected from available ones. Typically, definition of usage contexts and reasoning models heavily relies on domain knowledge. However, in practice many applications are used in so diverse situations that no developer can predict them all and collect for each situation adequate training and test databases. Such applications have to adapt to a new user or unknown context at runtime just from interaction with the user, preferably in fairly lightweight ways, that is, requiring limited user effort to collect training data and limited time of performing the adaptation. This paper analyses adaptation trends in several emerging domains and outlines promising ideas, proposed for making multimodal classifiers user-specific and context-specific without significant user efforts, detailed domain knowledge, and/or complete retraining of the classifiers. Based on this analysis, this paper identifies important application characteristics and presents guidelines to consider these characteristics in adaptation design.

show abstract

“…Human action retrieval is a different task to recognition, have only a single training sample and its results relies on the retrieval ranking [10]. In comparison to image retrieval, video retrieval has been attended less.…”

Section: Related Workmentioning

confidence: 99%

Content-based retrieval of human actions by analysing the statistical information of features

Ramezani

2014

2014 6th Conference on Information and Knowledge Technology (IKT)

View full text Add to dashboard Cite

Fast g rowth of video data on the Internet requires mana g in g these video data. In the last decade, Content-Based Video Retrieval (CBVR) became a considerable research interest to handle the lar g e amounts of collectin g video media on the Internet. Due to concernin g a lar g e ratio of the videos on the Internet to humans, human action retrieval is presented as a new topic in CBVR domain. In this paper, we seek to improve the current state-of-the-art retrieval al g orithms for CBVR by usin g the statistical information. The statistical information is utilized to represent the video by a vector with m units instead of local points of mXn units and creatin g a histo g ram of Ba g of Words and also, it decreases the complexity si g nificantly. Furthermore, each vector is used to compare the videos and to find the similar videos to the query one instead of histo g ram of Ba g of Words.The experimental results on KTH dataset illustrated that in contrast to the Ba g -of-Words model and its various parameters, our method can perform better.

show abstract

VisionGo: Towards video retrieval with joint exploration of human and computer

Cited by 18 publications

References 44 publications

Video Interaction Tools

Video Interaction Tools

Lightweight Adaptation of Classifiers to Users and Contexts: Trends of the Emerging Domain

Content-based retrieval of human actions by analysing the statistical information of features

Contact Info

Product

Resources

About