Proposing the EventDance framework for unsupervised cross-modal adaptation in event-based object recognition.
提案されたEventDanceフレームワークは、画像からイベントへの適応を実現し、先行研究を凌駕する性能を示す。
EventDance++ tackles the challenging problem of cross-modal (image-to-events) adaptation for event-based recognition without accessing any labeled source image data. It leverages language-guided reconstruction-based modality bridging and multi-representation knowledge adaptation to effectively bridge the modality gap and transfer knowledge from image to event domains.