DFKI and University of Kaiserslautern participation at TRECVID 2010 - Semantic Indexing Task

Run No. Run ID Run Description infMAP (%) training on IACC data 1 F A DFKI-MADM 3 SIFT visual words, Color Correlograms and Face-Detection separately trained, late fusion of SVMs scores 5.0 2 F A DFKI-MADM 4 SIFT visual words with SVMs 4.4 combined training on YouTube 3 F D DFKI-MADM 1 SIFT visual words, Color Correlograms and Face-Detection separately trained, late fusion of SVMs scores 2.1 4 F B DFKI-MADM 2 SIFT visual words with SVMs 1.3 This paper describes the TRECVID 2010 participation of the DFKI-MADM team in the semantic indexing task. This years participation was dominated by two aspects, a new dataset and a large-sized vocabulary of 130 concepts. For the annual TRECVID benchmark this means to scale label annotation efforts to significant larger concept vocabularies and datasets. Aiming to reduce this effort, our intention is to automatically acquire training data from online video portals like YouTube and to use tags, associated with each video, as concept labels. Results for the evaluated subset of concepts show similarly to last year’s participation [3], that effects like label noise and domain change lead to a performance loss (infMAP 2.1% and 1.3%) as compared to purely TRECVID trained concept detectors (infMAP 5.0% and 4.4%). Nevertheless, for individual concepts like “demonstration or protest” or “bus” automatic learning from online video portals is a valid alternative to expected labeled training datasets. Furthermore, the results also show that fusion of multiple features helps to improves detection precision.