Efficient Shape Matching Using Shape Contexts
IEEE Transactions on Pattern Analysis and Machine Intelligence
MULTIMEDIA '06 Proceedings of the 14th annual ACM international conference on Multimedia
Interest-based personalized search
ACM Transactions on Information Systems (TOIS)
VideoSense: towards effective online video advertising
Proceedings of the 15th international conference on Multimedia
Advertising keyword suggestion based on concept hierarchy
WSDM '08 Proceedings of the 2008 International Conference on Web Search and Data Mining
Argo: intelligent advertising made possible from users' photos
MM '09 Proceedings of the 17th ACM international conference on Multimedia
Semantic Linking between Video Ads and Web Services with Progressive Search
ICDMW '09 Proceedings of the 2009 IEEE International Conference on Data Mining Workshops
SURF: speeded up robust features
ECCV'06 Proceedings of the 9th European conference on Computer Vision - Volume Part I
Hi-index | 0.00 |
The rapid popularization of various online media services have attracted large amounts of consumers and shown us a large potential market of video advertising. In this paper, we aim to produce interactive service recommendation based on ad concept hierarchy by linking web videos, especially ad videos, with informative product details over the commercial websites. By introducing the domain based concept hierarchy, the recommendation quality is greatly improved. Given an ad video, we will try to semantically analyze it and conduct a contextual search from two aspects: video content and tags. For video content, we firstly extract its key frames and then make a visual search to find some relevant products. For video tags (if any) and relevant product tags gained by visual search, we will launch a textual search based on our ad concept hierarchy to judge the product category, generate some suggestion keywords, and give some recommended products to users. Users can also interactively select and adjust product categories and keywords to personalize their intentions by textual re-search. Our experimental results show that the system can successfully provide suggestion that meets the relevancy and individual requirements.