论文信息 - Towards Reliable Automatic Multimodal Content Analysis

Towards Reliable Automatic Multimodal Content Analysis

This poster presents a pilot where audio description is used to enhance automatic content analysis, for a project aiming at creating a tool for easy access to large AV archives.

Jorma Laaksonen | Mikko Kurimo | Maija Hirvonen | Liisa Tiittula | Olli Philippe Lautenbacher

[1] Pilar Orero,et al. Audio Description: New perspectives illustrated , 2014 .

[2] Mikko Kurimo,et al. Bounded Conditional Mean Imputation with Observation Uncertainties and Acoustic Model Adaptation , 2015, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[3] Jorma Laaksonen,et al. Advances in visual concept detection: Ten years of TRECVID , 2015 .

[4] Agnieszka Szarkowska,et al. Subtitles on the moving image: an overview of eye tracking studies , 2015 .

[5] Irene Ranzato,et al. Anna Maszerowska, Anna Matamala & Pilar Orero (eds.). (2014) Audio description. New perspectives illustrated. , 2016 .

[6] Bernt Schiele,et al. A dataset for Movie Description , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7] Fei-Fei Li,et al. Deep visual-semantic alignments for generating image descriptions , 2015, CVPR.

[8] Fei-Fei Li,et al. Deep visual-semantic alignments for generating image descriptions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).