Classifying visual and audio events of various durations in videos with MM-Pyramid

Classifying visual and audio events of various durations in videos with MM-Pyramid

MM-Pyramid: Multimodal Pyramid Attentional Network for Audio-Visual Event Localization and Video Parsing

arXiv paper abstract https://arxiv.org/abs/2111.12374v1

arXiv PDF paper https://arxiv.org/pdf/2111.12374v1.pdf

Recognizing and localizing events in videos is a fundamental task for video understanding ... events may occur in auditory and visual modalities

... Most previous works ... do not consider semantic information at multiple scales ... makes ... difficult to localize events

... present a Multimodal Pyramid Attentional Network (MM-Pyramid) ... integrates multi-level temporal features for audio-visual event localization and ... parsing.

... propose a novel attentive feature pyramid module, which is composed of the fixed-size attention mechanism and dilated convolution block.

... experiments on the AVE and LLP datasets demonstrate the effectiveness of our proposed approach on localizing events in multiple lengths.

Please like and share this post if you enjoyed it using the buttons at the bottom! Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact Web site with my other posts by category https://morrislee1234.wixsite.com/website #ComputerVision #Video #AINewsClips #AI #ML #ArtificialIntelligence #MachineLearning