Temporal Binary Representation for Event-Based Action Recognition
Authors:
- 10 January 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 10426-10432
Abstract
In this paper we present an event aggregation strategy to convert the output of an event camera into frames processable by traditional Computer Vision algorithms. The proposed method first generates sequences of intermediate binary representations, which are then losslessly transformed into a compact format by simply applying a binary-to-decimal conversion. This strategy allows us to encode temporal information directly into pixel values, which are then interpreted by deep learning models. We apply our strategy, called Temporal Binary Representation, to the task of Gesture Recognition, obtaining state of the art results on the popular DVS128 Gesture Dataset. To underline the effectiveness of the proposed method compared to existing ones, we also collect an extension of the dataset under more challenging conditions on which to perform experiments.Keywords
All Related Versions
Funding Information
- MIUR (20172BH297)
This publication has 30 references indexed in Scilit:
- Segmentation Free Object Discovery in Video2016
- Fast Temporal Activity Proposals for Efficient Detection of Human Actions in Untrimmed Videos2016
- Finding action tubes2015
- Going deeper with convolutions2015
- Temporal Localization of Actions with ActomsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2013
- Real-time classification and sensor fusion with a spiking deep belief networkFrontiers in Neuroscience, 2013
- Human activity analysisACM Computing Surveys, 2011
- A survey on vision-based human action recognitionImage and Vision Computing, 2010
- Gesture Recognition: A SurveyIEEE Transactions on Systems, Man and Cybernetics, Part C (Applications and Reviews), 2007
- OHAJIKI Interface: Flicking Gesture Recognition with a High-Speed CameraLecture Notes in Computer Science, 2006