ActivityNet Captions

AbstractGenerating multi-sentence descriptions for videos is one of the most challenging captioning tasks due to its high requirements for not only visual relevance but also discourse-based coherence across the sentences …...

AbstractTemporal language grounding in videos aims to localize the temporal span relevant to the given query sentence. Previous methods treat it either as a boundary regression task or a span …...

AbstractVideo sentence localization aims to locate moments in an unstructured video according to a given natural language query. A main challenge is the expensive annotation costs and the annotation bias. …...