• Corpus ID: 17459899

TRECVID 2003 Experiments at Media Team Oulu and VTT

  title={TRECVID 2003 Experiments at Media Team Oulu and VTT},
  author={Mika Rautiainen and Jani Penttil{\"a} and Paavo Pietarila and Kai Noponen and Matti Hosio and Timo Koskela and Satu-Marja M{\"a}kel{\"a} and Johannes Peltola and Jialin Liu and Timo Ojala and Tapio Sepp{\"a}nen and Mediateam Oulu},
MediaTeam Oulu and VTT Technical Research Centre of Finland participated jointly in semantic feature extraction, manual search and interactive search tasks of TRECVID 2003. We participated to the semantic feature extraction by submitting results to 15 out of the 17 defined semantic categories. Our approach utilized spatio-temporal visual features based on correlations of quantized gradient edges and color values together with several physical features from the audio signal. Most recent version… 
Analysing the performance of visual, concept and text features in content-based video retrieval
Weighted fusion of text, concept and visual features improved the performance over text search baseline, and expanded query term list of text queries gave also notable increase in performance over the baseline text search.
Assessing User Behaviour in News Video Retrieval
Analysis of the results at various stages in the retrieval process suggests that retrieval based on transcriptions of the speech in video data adds more to the average precision of the result than content-based image retrieval basedon low-level visual features.
Comparison of Visual Features and Fusion Techniques in Automatic Detection of Concepts from News Video
Experiments on automatic detection of semantic concepts, which are textual descriptions about the digital video content, show that the feature fusion based on ranked lists gives better detection performance than fusion of normalized low-level feature spaces distances.
Cluster-temporal browsing of large news video databases
Results indicate improvements in browsing efficiency when automatic speech recognition transcripts are incorporated into browsing by visual similarity, and performed well in overall comparison with interactive video retrieval systems in TRECVID 2003 evaluation.
On the detection of semantic concepts at TRECVID
Trends in the emerging concept detection systems, architectures and algorithms are studied and strategies that have yielded reasonable success, and challenges and gaps that lie ahead are analyzed.
Mediateam Oulu Background and Mission
MediaTeam conducts research on the features, use, and applications of multimedia and digital media types (image, sound, video, text) in information and communication systems. MediaTeam’s research
Semantic annotation for retrieval of visual resources
Laura Hollink onderzoekt de problemen bij het zoeken naar beeldmateriaal en de mogelijke oplossingen daarvoor, in drie uiteenlopende collecties: schilderijen, foto’s van organische cellen en nieuwsuitzendingen.


Temporal color correlograms for video retrieval
The efficiency of the temporal color correlogram and HSV color correlograms are evaluated against other retrieval systems participating the TREC video track evaluation and against color histograms used commonly in content-based retrieval.
The LIMSI Broadcast News transcription system
Development work in moving from laboratory read speech data to real-world or `found' speech data in preparation for the DARPA evaluations on this task from 1996 to 1999 is described.
Detecting Semantic Concepts from Video Using Temporal Gradients and Audio Classification
New methods to detect semantic concepts from digital video based on audible and visual content and Temporal Gradient Correlogram captures temporal correlations of gradient edge directions from sampled shot frames are described.
Semantic image retrieval with hsv correlograms
This work studies content-based retrieval of images using color correlograms computed in HSV color space, and tries to make the correlogram more sensitive to changes in color content and less sensitive to illumination by quantisizing the hue component more precisely than the value component.
In recent years the field of content-based audio signal classification and retrieval has gained a growing amount of interest among researchers around the world. This paper describes a technique,
Face Detection in Still Gray Images
A trainable system for detecting frontal and near-frontal views of faces in still gray images using Support Vector Machines (SVMs), and a component-based method for face detection consisting of a two-level hierarchy of SVM classifiers.
Sound onset detection by applying psychoacoustic knowledge
  • A. Klapuri
  • Computer Science
    1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258)
  • 1999
A system was designed, which is able to detect the perceptual onsets of sounds in acoustic signals and utilizes band-wise processing and a psychoacoustic model of intensity coding to combine the results from the separate frequency bands.
Real-time digital hardware pitch detector
Computing of the autocorrelation function of the clipped speech is easily implemented in digital hardware using simple combinatorial logic, i.e., an up-down counter can be used to compute each correlation point.
Decision Combination in Multiple Classifier Systems
This work proposes three methods based on the highest rank, the Borda count, and logistic regression for class set reranking that have been tested in applications of degraded machine-printed characters and works from large lexicons, resulting in substantial improvement in overall correctness.
Spectral analysis and discrimination by zero-crossings
  • B. Kedem
  • Computer Science
    Proceedings of the IEEE
  • 1986
The theme of this work is that higher order crossings analysis provides a useful descriptive as well as an analytical tool that can in many respects match spectral analysis.