Ad-Net: Audio-Visual Convolutional Neural Network for Advertisement Detection In Videos
This addresses the need for personalized advertising in online businesses and video broadcasting by automating ad detection, though it is incremental as it builds on existing audio-visual fusion methods.
The paper tackled the problem of detecting advertisements in videos to enable personalized ad replacement, proposing a two-stream audio-visual CNN that improved performance over hand-crafted feature models, trained on over 50k video and commercial shots.
Personalized advertisement is a crucial task for many of the online businesses and video broadcasters. Many of today's broadcasters use the same commercial for all customers, but as one can imagine different viewers have different interests and it seems reasonable to have customized commercial for different group of people, chosen based on their demographic features, and history. In this project, we propose a framework, which gets the broadcast videos, analyzes them, detects the commercial and replaces it with a more suitable commercial. We propose a two-stream audio-visual convolutional neural network, that one branch analyzes the visual information and the other one analyzes the audio information, and then the audio and visual embedding are fused together, and are used for commercial detection, and content categorization. We show that using both the visual and audio content of the videos significantly improves the model performance for video analysis. This network is trained on a dataset of more than 50k regular video and commercial shots, and achieved much better performance compared to the models based on hand-crafted features.