Recognising complex activities with histograms of relative tracklets
作者：
Highlights：
• We propose a method for activity recognition from video and accelerometer data.
• Visual accelerometer localization and tracking establishes cross-modal relations.
• RETLETS encode relative motion between tracked objects and local visual features.
• Recognition using various feature combinations is evaluated on the 50 Salads dataset.
• The method using RETLETS outperforms the state-of-the-art on this dataset.
摘要
•We propose a method for activity recognition from video and accelerometer data.•Visual accelerometer localization and tracking establishes cross-modal relations.•RETLETS encode relative motion between tracked objects and local visual features.•Recognition using various feature combinations is evaluated on the 50 Salads dataset.•The method using RETLETS outperforms the state-of-the-art on this dataset.
论文关键词：Activity recognition,Relative tracklets,Sensor fusion,Food preparation
论文评审过程：Received 10 March 2016, Revised 14 July 2016, Accepted 31 August 2016, Available online 1 September 2016, Version of Record 6 December 2016.
论文官网地址：https://doi.org/10.1016/j.cviu.2016.08.012
