Talking to Me or Someone Else? Rethinking Talk-to-Me Detection in Egocentric Videos
This paper reformulates TTM as an online, frame-level prediction task, instead of treating TTM as a binary problem against a single negative class, and model it in the presence of diverse and previously underexplored non-TTM states, such as talking-to-others, self-talking, and background conditions.
Fei-Yu Du, Xi He, Jia Li et al.
· 0 citations