Advances in Multimedia Modeling: 14th International Multimedia Modeling Conference, MMM 2008, Kyoto, Japan, January 9-11, 2008, ProceedingsShin'ichi Satoh, Frank Nack, Minoru Etoh Springer, 2008. gada 22. janv. - 512 lappuses Welcometothe14thInternationalMultimediaModelingConference(MMM2008), held January 9–11, 2008 at Kyoto University, Kyoto, Japan. MMM is a leading international conference for researchersand industry practitioners to share their new ideas, original research results and practical development experiences from all multimedia related areas. It was a great honor to have MMM2008, one of the most long-standing m- timedia conferences, at one of the most beautiful and historically important Japanese cities. Kyoto was an ancient capital of Japan, and was and still is at the heartofJapanesecultureandhistory. Kyotoinwintermaydistinctivelyo?er the sober atmosphere of an ink painting. You can enjoy old shrines and temples which are designated as World Heritage Sites. The conference venue was the Clock Tower Centennial Hall in Kyoto University, which is one of the oldest universities in Japan. MMM2008 featured a comprehensive program including three keynote talks, six oral presentation sessions, and two poster and demo sessions. The 133 s- missions included a large number of high-quality papers in multimedia content analysis, multimedia signal processing and communications, and multimedia applications and services. We thank our 137 Technical Program Committee members and reviewers who spent many hours reviewing papers and prov- ing valuable feedback to the authors. Based on the 3 or 4 reviews per paper the Program Chairs decided to accept only 23 as oral papers and 24 as poster papers, where each type of presentation could in addition present the work as a demo. The acceptance rate of 36% follows the MMM tradition of ful?lling fruitful discussions throughout the conference. |
No grāmatas satura
1.–5. rezultāts no 36.
15. lappuse
... labeled images from each class as training dataset to select a set of highly informative and distinctive visual words based on Information Gain criteria [14]. It then takes the joint probabilities of visual words and image Phase 1: Bag ...
... labeled images from each class as training dataset to select a set of highly informative and distinctive visual words based on Information Gain criteria [14]. It then takes the joint probabilities of visual words and image Phase 1: Bag ...
16. lappuse
... labeled training set) as input to perform sequential Information Bottleneck (sIB) clustering algorithm [13] to group visual words with similar class probability distributions into visual synsets. According to the clustering assignment ...
... labeled training set) as input to perform sequential Information Bottleneck (sIB) clustering algorithm [13] to group visual words with similar class probability distributions into visual synsets. According to the clustering assignment ...
25. lappuse
... labeling, the only instruction to the sub-layer labeling is implicitly given through the MI relation. Due to multi-layer extension, the ambiguity in MI setting will propagate layer by layer, we define such phenomena as ambiguity ...
... labeling, the only instruction to the sub-layer labeling is implicitly given through the MI relation. Due to multi-layer extension, the ambiguity in MI setting will propagate layer by layer, we define such phenomena as ambiguity ...
28. lappuse
... labeled noisy data, and learn the hyper-plane more effectively. Accordingly, it is desirable to utilize the kernel between sub-structures for MLMI learning, and additional explicit constraint reflecting multi-instance relation and inter ...
... labeled noisy data, and learn the hyper-plane more effectively. Accordingly, it is desirable to utilize the kernel between sub-structures for MLMI learning, and additional explicit constraint reflecting multi-instance relation and inter ...
29. lappuse
... labeled as +1, and negative ones being –1, by multiinstance relationship, the prediction of hyper-bag f(T) should be consistent with the maximal prediction of its Ith sub-layer maxm_1,...,N. {f(T,)}. Various loss functions could be ...
... labeled as +1, and negative ones being –1, by multiinstance relationship, the prediction of hyper-bag f(T) should be consistent with the maximal prediction of its Ith sub-layer maxm_1,...,N. {f(T,)}. Various loss functions could be ...
Saturs
1 | |
13 | |
24 | |
35 | |
New Approach for Hierarchical Classifier Training and Multilevel Image Annotation | 45 |
Extracting Text Information for ContentBased Video Retrieval | 58 |
RealTime Video Surveillance Based on Combining Foreground Extraction and Human Detection | 70 |
Detecting and Clustering Multiple Takes of One Scene | 80 |
Blurred Image Detection and Classification | 277 |
CrossLingual Retrieval of Identical News Events by NearDuplicate Video Segment Detection | 287 |
Web Image Gathering with a PartBased Object Recognition Method | 297 |
A Query Language Combining Object Features and Semantic Events for Surveillance Video Retrieval | 307 |
Semantic Quantization of 3D Human Motion Capture Data Through SpatialTemporal Feature Extraction | 318 |
Fast Intermode Decision Via Statistical Learning for H264 Video Coding | 329 |
A Novel Motion Estimation Method Based on Normalized Cross Correlation for Video Compression | 338 |
Curved RayCasting for Displacement Mapping in the GPU | 348 |
An ImagesBased 3D Model Retrieval Approach | 90 |
Oh Web Image Where Art Thou? | 101 |
Complementary Variance Energy for Fingerprint Segmentation | 113 |
Similarity Search in Multimedia Time Series Data Using AmplitudeLevel Features | 123 |
Sound Source Localization with Noncalibrated Microphones | 134 |
Privacy Protected Video Surveillance System Using Adaptive Visual Abstraction | 144 |
DistributionBased Similarity for Multirepresented Multimedia Objects | 155 |
A Multimodal Input Device for Music Authoring for Children | 165 |
FreeShaped Video Collage | 175 |
AestheticsBased Automatic Home Video Skimming System | 186 |
Using Fuzzy Lists for Playlist Management | 198 |
Tagging Video Contents with PositiveNegative Interest Based on Users Facial Expression | 210 |
A MixedReality Game Based on Scene Identification | 220 |
RealTime Multiview Object Tracking in Mediated Environments | 230 |
Reconstruct 3D Human Motion from Monocular Video Using Motion Library | 242 |
Appropriate Segment Extraction from Shots Based on Temporal Patterns of Example Videos | 253 |
Fast Segmentation of H264AVC Bitstreams for OnDemand Video Summarization | 265 |
EmotionBased Music Visualization Using Photos | 358 |
Distant Collaboration Support System for Manufacturers | 369 |
Accurate Identifying Method of JPEG2000 Images for Digital Cinema | 380 |
Optimization of Spatial Error Concealment for H264 Featuring Low Complexity | 391 |
Temporal Error Concealment for H264 Using Optimum Regression Plane | 402 |
Transform Domain WynerZiv Codec Based on Turbo Trellis Codes Modulation | 413 |
Selective Sampling Based on Dynamic Certainty Propagation for Image Retrieval | 425 |
Local Radon Transform and Earth Movers Distances for ContentBased Image Retrieval | 436 |
Content Based Querying and Searching for 3D Human Motions | 446 |
Bimodal Conceptual Indexing for Medical Image Retrieval | 456 |
Audio Analysis for Multimedia Retrieval from a Ubiquitous Home | 466 |
Effectiveness of Signal Segmentation for Music Content Representation | 477 |
Probabilistic Estimation of a Novel Music Emotion Model | 487 |
Embedding High BitRate Data in Audio | 498 |
Author Index | 508 |
Citi izdevumi - Skatīt visu
Advances in Multimedia Modeling: 14th International Multimedia Modeling ... Shin'ichi Satoh,Frank Nack,Minoru Etoh Ierobežota priekšskatīšana - 2007 |
Bieži izmantoti vārdi un frāzes
AdaBoost algorithm analysis applied approach audio automatic average Berlin Heidelberg 2008 bitplane block blurred camera classifier clips clustering codec coding coefficients collage color color histogram component Computer Computer Vision content-based image retrieval corresponding database decoder defined denotes detection distance distribution emotion encoder error concealment estimation Etoh Eds evaluation feature extraction feature vectors feedback filtering frame function fuzzy lists Google Images Heidelberg histogram IEEE IEEE Trans image concepts image retrieval indexing input kernel keywords labeled learning LightCollabo LNCS macroblocks mapping matching microphones motion capture Multimedia object OFDM paper parameters partition performance pixels playlist Proc proposed method PSNR quantization query Radon Transform region relevant representation sampling scene Section selection semantic sequence shot signal similarity space spatial support vector machines synset techniques texture threshold tracking transform TRECVID users visual features visual synset visual words