5,693 Hits in 4.5 sec

Few-Shot Abstract Visual Reasoning With Spectral Features [article]

Tanner Bohn, Yining Hu, Charles X. Ling
2019 arXiv   pre-print
We present an image preprocessing technique capable of improving the performance of few-shot classifiers on abstract visual reasoning tasks.  ...  Many visual reasoning tasks with abstract features are easy for humans to learn with few examples but very difficult for computer vision approaches with the same number of samples, despite the ability  ...  There are many potential sources of abstract visual reasoning tasks one could study.  ... 
arXiv:1910.01833v1 fatcat:gtptrs4dk5gjvf7o5vkp73eknu

Hyperspectral Image Classification across Different Datasets: A Generalization to Unseen Categories

Erting Pan, Yong Ma, Fan Fan, Xiaoguang Mei, Jun Huang
2021 Remote Sensing  
In this work, we utilize a three-phase scheme, including feature embedding, feature mapping, and label reasoning.  ...  Extensive experiments on two pairs of datasets with different comparative methods have shown the effectiveness and potential of zero-shot learning in HSI classification.  ...  Learn h(·) on T R with label y tr i :X →Z 7 case Semantic-Visual Mapping do 8 Learn h(·) on T R with label y tr i :Z →X 9 end 10 Phase C: Label Reasoning 11 for ∀x te j ∈ X T E do 12 Get the hyperspectral  ... 
doi:10.3390/rs13091672 doaj:61b62fe19af84a67af06cfa743ff71ad fatcat:oem53qkonbhcrfjodti5hwpege

Improving acoustic speaker verification with visual Body-Language features

Christoph Bregler, George Williams, Sally Rosenthal, Ian McDowall
2009 2009 IEEE International Conference on Acoustics, Speech and Signal Processing  
We show how an SVM based acoustic speaker verification system can be significantly improved in incorporating new visual features that capture the speaker's "Body Language."  ...  With shots that are longer then 5 minutes (i.e. a speech), our shot-detector cuts the video into 5 minute shots. Sometimes we get very short shots of just a few seconds.  ...  Introduction Among the reasons for recent advances in speaker recognition are discriminative classification based on SVMs [10] and novel feature extraction methods, based on both short-term spectral  ... 
doi:10.1109/icassp.2009.4959982 dblp:conf/icassp/BreglerWRM09 fatcat:dlkvpujusjfnnpwpl4bl7kl2ke

Content-based representation and retrieval of visual media: A state-of-the-art review

Philippe Aigrain, Hongjiang Zhang, Dragutin Petkovic
1996 Multimedia tools and applications  
This paper reviews a number of recently available techniques in content analysis of visual media and their application to the Indexing, retrieval, abstracting, relevance assessment, interactive perception  ...  , annotation and re-use of visual documents.  ...  There are many different forms of sequences, from a field-counterfield sequence in motion picture (with many repetitions of 2 shot types interlaced with a few other shot types) to a sequence of 2 shots  ... 
doi:10.1007/bf00393937 fatcat:xrxszt23qrb6pci5iqo7ybmshe

Group-based spatio-temporal video analysis and abstraction using wavelet parameters

M. Omidyeganeh, S. Ghaemmaghami, S. Shirmohammadi
2011 Signal, Image and Video Processing  
Our contribution is that this still image video abstraction scheme does not need shot or cluster boundary detection, unlike current methods.  ...  1 In this paper, we present a spatio-temporal event based approach to video signal analysis and abstraction employing wavelet transform features.  ...  This is while each visual event can be assumed to be independent of other visual events, considering the reasonable assumption that the video director has already thought about the occurrence of each of  ... 
doi:10.1007/s11760-011-0268-y fatcat:2qmwyihjsza3himiv7gwhdtulu

Deep Relation Network for Hyperspectral Image Few-Shot Classification

Kuiliang Gao, Bing Liu, Xuchu Yu, Jinchun Qin, Pengqiang Zhang, Xiong Tan
2020 Remote Sensing  
This paper aims to explore how to accurately classify new hyperspectral images with only a few labeled samples, i.e., the hyperspectral images few-shot classification.  ...  Firstly, the feature learning module and the relation learning module of the model can make full use of the spatial–spectral information in hyperspectral images and carry out relation learning by comparing  ...  The authors would also like to thank all the professionals for kindly providing the codes associated with the experiments. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/rs12060923 fatcat:vxm6xzqjk5dftj73zn2mq5edka

Organizing Multimedia Information with Maps [chapter]

Thomas Bärecke, Ewa Kijak, Marcin Detyniecki, Andreas Nürnberger
2008 Studies in Computational Intelligence  
.: Organizing Multimedia Information with Maps, Studies in Computational Intelligence (SCI) 96, 493-509 (2008)  ...  We introduce a novel time bar visualization that reprojects the temporal information.  ...  Feature Extraction In order to obtain a good clustering, a reasonable representation of the video segments is necessary.  ... 
doi:10.1007/978-3-540-76827-2_18 fatcat:zqjksuhlz5cvxngdykhykxwf7y

Automatic analysis of movies for content characterization

Sher Muhammad Doudpota, Sumanta Guha
2010 2010 International Conference on Networking and Information Technology  
Every scene in a movie has certain audio-visual features that differentiate it from other scenes.  ...  On other hand, an action scene might have rapid movements, lot of sound effects and short shot duration with less or no repetition of shots.  ... 
doi:10.1109/icnit.2010.5508472 fatcat:cctmwvvidrfkzmlpyt3v64ptzm

Cross-Modal Analysis of Audio-Visual Film Montage

Matthias Zeppelzauer, Dalibor Mitrovic, Christian Breiteneder
2011 2011 Proceedings of 20th International Conference on Computer Communications and Networks (ICCCN)  
We propose a cross-modal approach that extracts sequences from a movie with synchronous audio-visual montage. Experiments confirm that the extracted sequences have high semantic relevance.  ...  Consequently, they represent a useful basis for different high-level movie abstraction tasks such as automated movie annotation and movie summarization.  ...  However, we observe that this assumption is not sufficient for the detection of synchronous montage in feature films for two reasons.  ... 
doi:10.1109/icccn.2011.6005782 dblp:conf/icccn/ZeppelzauerMB11 fatcat:bijgniljsfew5gtuorbezjdaxe

Hierarchical Video Summaries By Dendrogram Cluster Analysis

Sergio Benini, Aldo Bianchetti, R. Leonardi, P. Migliorati
2006 Zenodo  
The proposed procedure can employ any visual low-level feature provided with a method to estimate similarity between shots.  ...  LOW-LEVEL COLOR FEATURES The proposed procedure for the creation of hierarchical summaries can employ any visual low-level feature provided with a method to estimate a shot-to-shot similarity.  ... 
doi:10.5281/zenodo.53330 fatcat:jfndit5vuzgu3biloc5lenga5e

Video Summarization: Survey on Event Detection and Summarization in Soccer Videos

Yasmin S., Soudamini Pawar
2015 International Journal of Advanced Computer Science and Applications  
We have discussed some features used for generating video summaries. As soccer is the world's most famous game played and watched, it is taken as a case study.  ...  In such cases, the users may just want to view the summary of the video that is just an abstract of the original video, instead of watching the whole video that provides more information about the occurrence  ...  Compared with static summarization, there are relatively few works being addressed for dynamic video skimming. Most techniques are based mainly on visual information.  ... 
doi:10.14569/ijacsa.2015.061133 fatcat:pvjtxbtwwfevzjt4d4s2q4rdly

Video genre categorization and representation using audio-visual information

Bogdan Ionescu
2012 Journal of Electronic Imaging (JEI)  
We propose an audio-visual approach to video genre classification using content descriptors that exploit audio, color, temporal, and contour information.  ...  Finally, we discuss a 3D video browsing platform that displays movies using feature-based coordinates and thus regroups them according to genre.  ...  has been supported by the Sectoral Operational Programme Human Resources Develop- The authors would like to thank CITIA -The City of Moving Images and Folimage Animation Company for providing them with  ... 
doi:10.1117/1.jei.21.2.023017 fatcat:ftmpjzlx5rdndbcmwiqx7jktga

Mapping Large-Scale Plateau Forest in Sanjiangyuan Using High-Resolution Satellite Imagery and Few-Shot Learning

Zhihao Wei, Kebin Jia, Xiaowei Jia, Pengyu Liu, Ying Ma, Ting Chen, Guilian Feng
2022 Remote Sensing  
We then propose an few-shot learning method for mapping plateau forests.  ...  The proposed few-shot learning method reached an F1-score of 84.23%, and outperformed the state-of-the-art object segmentation methods.  ...  Experiments show that our proposed few-shot learning method outperforms several state-of-the-art algorithms.  ... 
doi:10.3390/rs14020388 fatcat:3pgbc6csuzbqjei4ign6kaqh3q

Multimodal and ontology-based fusion approaches of audio and visual processing for violence detection in movies

Thanassis Perperis, Theodoros Giannakopoulos, Alexandros Makris, Dimitrios I. Kosmopoulos, Sofia Tsekeridou, Stavros J. Perantonis, Sergios Theodoridis
2011 Expert systems with applications  
one, that combines the audio-visual cues with violence and multimedia ontologies.  ...  Towards this goal, a multi-step approach is followed: initially, automated audio and visual analysis is performed to extract audio and visual cues.  ...  Few attempts appear in the literature for semantic video analysis in more abstract domains.  ... 
doi:10.1016/j.eswa.2011.04.219 fatcat:vfvplzidkzhrnhek7h3kroi5ia

Discriminative genre-independent audio-visual scene change detection

Kevin W. Wilson, Ajay Divakaran, Raimondo Schettini, Ramesh C. Jain, Simone Santini
2009 Multimedia Content Access: Algorithms and Systems III  
ABSTRACT We present a technique for genre-independent scene-change detection using audio and video features in a discriminative support vector machine (SVM) framework.  ...  We also find that the genres that benefit the most are those with which the previous audio-only was least effective.  ...  The color bhattacharyya feature by itself is substantially better than our previous visual feature based on shot changes, and the combination of shot change, color bhattacharyya, and audio features yields  ... 
doi:10.1117/12.805624 fatcat:d3wbtf32hzhz5hvl5uwtyxta3y
« Previous Showing results 1 — 15 out of 5,693 results