Past and Future Motion Guided Network for Audio Visual Event Localization
This work addresses the problem of localizing events in videos for multimedia analysis, but it is incremental as it builds on existing methods by adding motion correlation.
The paper tackles audio-visual event localization by proposing a network that incorporates past and future motion to guide attention, achieving state-of-the-art performance on the AVE dataset in both supervised and weakly-supervised settings.
In recent years, audio-visual event localization has attracted much attention. It's purpose is to detect the segment containing audio-visual events and recognize the event category from untrimmed videos. Existing methods use audio-guided visual attention to lead the model pay attention to the spatial area of the ongoing event, devoting to the correlation between audio and visual information but ignoring the correlation between audio and spatial motion. We propose a past and future motion extraction (pf-ME) module to mine the visual motion from videos ,embedded into the past and future motion guided network (PFAGN), and motion guided audio attention (MGAA) module to achieve focusing on the information related to interesting events in audio modality through the past and future visual motion. We choose AVE as the experimental verification dataset and the experiments show that our method outperforms the state-of-the-arts in both supervised and weakly-supervised settings.