Online Action Detection
In online action detection, the goal is to detect the start of an action in a video stream as soon as it happens. For instance, if a child is chasing a ball, an autonomous car should recognize what is going on and respond immediately. This is a very chall
- PDF / 3,081,241 Bytes
- 16 Pages / 439.37 x 666.142 pts Page_size
- 57 Downloads / 194 Views
ESAT - PSI, KU Leuven, Leuven, Belgium {roeland.degeest,amir.ghodrati,tinne.tuytelaars}@esat.kuleuven.be 2 QUVA, University of Amsterdam, Amsterdam, Netherlands {e.gavves,z.li2,c.g.m.snoek}@uva.n
Abstract. In online action detection, the goal is to detect the start of an action in a video stream as soon as it happens. For instance, if a child is chasing a ball, an autonomous car should recognize what is going on and respond immediately. This is a very challenging problem for four reasons. First, only partial actions are observed. Second, there is a large variability in negative data. Third, the start of the action is unknown, so it is unclear over what time window the information should be integrated. Finally, in real world data, large within-class variability exists. This problem has been addressed before, but only to some extent. Our contributions to online action detection are threefold. First, we introduce a realistic dataset composed of 27 episodes from 6 popular TV series. The dataset spans over 16 h of footage annotated with 30 action classes, totaling 6,231 action instances. Second, we analyze and compare various baseline methods, showing this is a challenging problem for which none of the methods provides a good solution. Third, we analyze the change in performance when there is a variation in viewpoint, occlusion, truncation, etc. We introduce an evaluation protocol for fair comparison. The dataset, the baselines and the models will all be made publicly available to encourage (much needed) further research on online action detection on realistic data.
Keywords: Action recognition
1
· Evaluation · Online action detection
Introduction
In this paper, we focus on the problem of online action detection. Unlike traditional action recognition and action detection as studied in the literature to date, e.g., [1–6], the goal of online action detection is to detect an action as it happens and ideally even before the action is fully completed. Being able to This work was supported by the KU Leuven GOA project CAMETRON. Electronic supplementary material The online version of this chapter (doi:10. 1007/978-3-319-46454-1 17) contains supplementary material, which is available to authorized users. c Springer International Publishing AG 2016 B. Leibe et al. (Eds.): ECCV 2016, Part V, LNCS 9909, pp. 269–284, 2016. DOI: 10.1007/978-3-319-46454-1 17
270
R. De Geest et al.
detect an action at the time of the occurence can be useful in many practical applications - think of a pro-active robot offering a helping hand; a surveillance camera raising an alarm not just after the facts but well in time to allow for intervention; a smart active camera system zooming in on the action scene and recording it from the optimal perspective; or an autonomous car stopping for a child chasing a ball (see Fig. 1). A similar task coined ‘early event detection’ has been brought to the attention of the community in the seminal work of Hoai and De la Torre [7,8]. However, they consider only the special case of relatively short video fra
Data Loading...
