Topic modelling for routine discovery from egocentric photo-streams

作者：

Highlights：

• We introduce a novel automatic unsupervised pipeline for the identification and characterization of Routine-related days from egocentric photo-streams.

• We prove that topic modelling is a powerful tool for discovery of patterns when addressing Bag-of-Words representation of photo-streams.

• We prove that using Dynamic-Time-Warping and Distance-based clustering is a robust technique to detect the cluster of routine days where the method is tolerant to small temporal differences in the daily events.

• We present and new egocentric dataset composed of a total of 100.000 images, from 104 days.

摘要

•We introduce a novel automatic unsupervised pipeline for the identification and characterization of Routine-related days from egocentric photo-streams.•We prove that topic modelling is a powerful tool for discovery of patterns when addressing Bag-of-Words representation of photo-streams.•We prove that using Dynamic-Time-Warping and Distance-based clustering is a robust technique to detect the cluster of routine days where the method is tolerant to small temporal differences in the daily events.•We present and new egocentric dataset composed of a total of 100.000 images, from 104 days.

论文关键词：Routine,Egocentric vision,Lifestyle,Behaviour analysis,Topic modelling

论文评审过程：Received 17 September 2019, Revised 28 February 2020, Accepted 12 March 2020, Available online 19 March 2020, Version of Record 31 March 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107330