AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions - Archive ouverte HAL Access content directly
Conference Papers Year : 2018

AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions

(1) , (1) , (1) , (1) , (1) , (1) , (1) , (1) , (1) , (1) , (1, 2) , (1, 3)
1
2
3
Chunhui Gu
  • Function : Author
Chen Sun
  • Function : Author
  • PersonId : 1004146
David Ross
Carl Vondrick
  • Function : Author
Caroline Pantofaru
  • Function : Author
Yeqing Li
  • Function : Author
George Toderici
  • Function : Author
Susanna Ricco
  • Function : Author
Rahul R Sukthankar
  • Function : Author

Abstract

This paper introduces a video dataset of spatio-temporally localized Atomic Visual Actions (AVA). The AVA dataset densely annotates 80 atomic visual actions in 437 15-minute video clips, where actions are localized in space and time, resulting in 1.59M action labels with multiple labels per person occurring frequently. The key characteristics of our dataset are: (1) the definition of atomic visual actions, rather than composite actions; (2) precise spatio-temporal annotations with possibly multiple annotations for each person; (3) exhaustive annotation of these atomic actions over 15-minute video clips; (4) people temporally linked across consecutive segments; and (5) using movies to gather a varied set of action representations. This departs from existing datasets for spatio-temporal action recognition, which typically provide sparse annotations for composite actions in short video clips. AVA, with its realistic scene and action complexity, exposes the intrinsic difficulty of action recognition. To benchmark this, we present a novel approach for action local-ization that builds upon the current state-of-the-art methods , and demonstrates better performance on JHMDB and UCF101-24 categories. While setting a new state of the art on existing datasets, the overall results on AVA are low at 15.8% mAP, underscoring the need for developing new approaches for video understanding.
Fichier principal
Vignette du fichier
AVA_CVPR2018_Camera_Ready.pdf (5.54 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01764300 , version 1 (11-04-2018)

Identifiers

Cite

Chunhui Gu, Chen Sun, David Ross, Carl Vondrick, Caroline Pantofaru, et al.. AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions. CVPR 2018 - Computer Vision and Pattern Recognition, Jun 2018, Salt Lake City, United States. pp.6047-6056, ⟨10.1109/CVPR.2018.00633⟩. ⟨hal-01764300⟩
512 View
277 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More