Finding action tubes
Citations Over TimeTop 1% of 2015 papers
Abstract
We address the problem of action detection in videos. Driven by the latest progress in object detection from 2D images, we build action models using rich feature hierarchies derived from shape and kinematic cues. We incorporate appearance and motion in two ways. First, starting from image region proposals we select those that are motion salient and thus are more likely to contain the action. This leads to a significant reduction in the number of regions being processed and allows for faster computations. Second, we extract spatio-temporal feature representations to build strong classifiers using Convolutional Neural Networks. We link our predictions to produce detections consistent in time, which we call action tubes. We show that our approach outperforms other techniques in the task of action detection.
Related Papers
- → Kinematic performance of planar 5R symmetrical parallel mechanism subjected to clearances and uncertainties(2018)21 cited
- → The Impact of the Support System’s Kinematic Structure on Selected Kinematic and Dynamic Quantities of an Experimental Crane(2014)7 cited
- Keeping Conflicts Latent: "Salient" versus "Non-Salient" Interpersonal Conflict Management Strategies of Japanese(2013)
- Detecting Salient Regions Based on "What" and "Where" Pathways of Visual Systems(2006)
- → A Computational Method to Find Salient Features(2008)