Disclaimer
If you're planing to use information provided on this site, please keep in mind that all numbers and papers are added by authors without double checking. We of course try to keep results as accurate as possible, and whenever we got notice of an error it will be fixed, but this does not release you from the obligation of reading the papers and double checking the numbers listed here before using them.

AVA

Dataset URL

Description : AVA is a dataset for spatio-temporal localization of atomic visual actions. The vocabulary consists of 80 different atomic visual actions. The dataset consists of 57.6k video segments collected from 192 different movies, where segments are 3 second long videos extracted sequentially in 15 minute chunks from each movie. Using chunks of 15 minutes per video enables diversity at the same time as continuity. A total of 210k actions are labeld. (see https://research.google.com/ava/ and http://thoth.inrialpes.fr/ava/ )

Number of Videos : 57600

Number of Classes : 80

Resources

Evaluation: MAP@IOU of 0.5 (test set)

Description: Mean average precision at IOU threshold of 0.5 on the test set

Results


Result Paper Description URL Peer Reviewed Year
Result Paper Description URL Peer Reviewed Year
18.4 AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions[Chunhui Gu, Chen Sun, Sudheendra Vijayanarasimhan, Caroline Pantofaru, David A. Ross, George Toderici, Yeqing Li, Susanna Ricco, Rahul Sukthankar, Cordelia Schmid, Jitendra Malik] RGB + Deep Flow URL No 2016
18.1 AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions[Chunhui Gu, Chen Sun, Sudheendra Vijayanarasimhan, Caroline Pantofaru, David A. Ross, George Toderici, Yeqing Li, Susanna Ricco, Rahul Sukthankar, Cordelia Schmid, Jitendra Malik] RGB + TVL-1 URL No 2016
27.2 Long-Term Feature Banks for Detailed Video Understanding[Chao-Yuan Wu, Christoph Feichtenhofer, Haoqi Fan, Kaiming He, Philipp Krähenbühl, Ross Girshick] LFB-NL (on test set) URL No 2018
27.1 SlowFast Networks for Video Recognition[Christoph Feichtenhofer, Haoqi Fan, Jitendra Malik, Kaiming He] SlowFast, +NL URL No 2018
21.91 A Better Baseline for AVA[Rohit Girdhar, João Carreira, Carl Doersch, Andrew Zisserman] I3D + FRCNN + JFT (RGB only) URL No 2018
24.93 Video Action Transformer Network[Rohit Girdhar, Joao Carreira, Carl Doersch, Andrew Zisserman] Tx+I3D+96f (RGB only) URL No 2019

If you want to add this result data into your web page, please insert the following HTML code on your web page:

Evaluation: MAP@IOU of 0.5 (val)

Description: MAP@IOU of 0.5 for the validation set

Results


Result Paper Description URL Peer Reviewed Year
Result Paper Description URL Peer Reviewed Year
27.7 Long-Term Feature Banks for Detailed Video Understanding[Chao-Yuan Wu, Christoph Feichtenhofer, Haoqi Fan, Kaiming He, Philipp Krähenbühl, Ross Girshick] LFB-NL (on validation set) URL No 2018
28.3 SlowFast Networks for Video Recognition[Christoph Feichtenhofer, Haoqi Fan, Jitendra Malik, Kaiming He] SlowFast++, +NL URL No 2018
22.8 A Better Baseline for AVA[Rohit Girdhar, João Carreira, Carl Doersch, Andrew Zisserman] I3D + FRCNN + JFT (RGB only) URL No 2018
25 Video Action Transformer Network[Rohit Girdhar, Joao Carreira, Carl Doersch, Andrew Zisserman] Tx+I3D+96f (RGB only) URL No 2019
20.4 Relational Action Forecasting[Chen Sun, Abhinav Shrivastava, Carl Vondrick, Rahul Sukthankar, Kevin Murphy, Cordelia Schmid] DR2N URL No 2019
23.6 Hierarchical Graph-Rnns for Action Detection of Multiple Activities[Sovan Biswas, Yaser Souri, Juergen Gall] I3D+Graph RNN URL No 2019

If you want to add this result data into your web page, please insert the following HTML code on your web page: