ActivityNet: A large-scale video benchmark for human activity understanding

Citations Over TimeTop 1% of 2015 papers

Abstract

In spite of many dataset efforts for human action recognition, current computer vision algorithms are still severely limited in terms of the variability and complexity of the actions that they can recognize. This is in part due to the simplicity of current benchmarks, which mostly focus on simple actions and movements occurring on manually trimmed videos. In this paper we introduce ActivityNet, a new large-scale video benchmark for human activity understanding. Our benchmark aims at covering a wide range of complex human activities that are of interest to people in their daily living. In its current version, ActivityNet provides samples from 203 activity classes with an average of 137 untrimmed videos per class and 1.41 activity instances per video, for a total of 849 video hours. We illustrate three scenarios in which ActivityNet can be used to compare algorithms for human activity understanding: untrimmed video classification, trimmed activity classification and activity detection.

Related Papers

• → Timely daily activity recognition from headmost sensor events(2019)63 cited
• → Activity Feature Solving Based on TF‐IDF for Activity Recognition in Smart Homes(2019)40 cited
• → Activity Recognition Using Hierarchical Hidden Markov Models on Streaming Sensor Data(2018)9 cited
• → Real-time recognition of activity levels for ambient assisted living(2015)5 cited
• → A user activity pattern mining system based on human activity recognition and location service(2018)6 cited