AdOn: toward contextual overlay in-video advertising

Abstract

This paper presents a contextual video advertising system, called AdOn, which supports intelligent overlay in-video advertising. Unlike most current ad-networks such as Youtube that overlay the ads at fixed locations in the videos (e.g., on the bottom fifth of videos 15 s in), AdOn is able to automatically detect a set of spatiotemporal non-intrusive locations and associate the contextually relevant ads with these locations. The overlay ad locations are obtained on the basis of video structuring, face and text detection, as well as visual saliency analysis, so that the intrusiveness to the users can be minimized. The ads are selected according to content-based multimodal relevance so that the relevance can be maximized. AdOn represents one of the first attempts towards contextual overlay video advertising by leveraging information retrieval and multimedia content analysis techniques. The experiments conducted on a video database with more than 100 video programs and 7,000 ad products indicated that AdOn is superior to existing advertising approaches in terms of ad relevance and user experience.

@ICT: attention-based virtual content insertion

“…Other methods tackle this problem from the viewpoint of visual attention. ImageSense [20], AdOn [22], and GameSence [12] insert the advertisements at the non-salient corner or side regions to avoid occluding the informative content of the image. In our previous work [16], a notation of lower attentive region is proposed and defined, from the cognitive point of view, as a region of the video frame which attracts less audience attention.…”

“…It needs little domain knowledge and is applied to general videos, such as TV play series and movies, etc. The basic idea of our overlay insertion method is similar to the one of AdOn [22]. Compared with AdOn, our work has the following characteristics:…”

“…Visual attention analysis is a rational way to generalize virtual content insertion. AdOn [22] is an instance of visual attention-based generic contextual in-video advertising system. It chooses attractive shots as insertion time, less attentive region as insertion place, and chooses advertisement according to textual relevance, user preference, and visual content consistency.…”

AdVisual

“…AdOn [4] and Videosense [5] mainly rely on the textbased strategy, which associate relevant ads with local subtitles and the descriptive texts on the web page. It is a fast and reasonable solution but cannot analyze local visual content.…”

Towards cost-driven control and management for cloud centric media network

“…B o ≤ B t : When the user is consuming the content on a bigger screen (e.g., TV), B o would be smaller than the retargeted stream size B t . In this case, the cloud clone inserts a few personalized video advertisements, by first picking the related advertisement video based on recommendation algorithms, then combining the selected video ad overlay with the targeted quality to the original content [96]. Finally, the combined video streaming is delivered to the end user.…”