THUMOS'13 Action Recognition Challenge Results:
Overall Ranking:
| Rank | Submission | Overall Accuracy | Split 1 Acc. | Split 2 Acc. | Split 3 Acc. |
| 1 | ID39_INRIA | 85.900% | 84.734% | 85.862% | 87.105% |
| 2 | ID40_Florence | 85.708% | 85.319% | 86.642% | 85.164% |
| 3 | ID35_Canberra | 85.437% | 84.761% | 86.367% | 85.183% |
| 4 | ID38_CAS_SIAT | 84.164% | 83.515% | 84.607% | 84.368% |
| 5 | ID25_Nanjing | 83.979% | 83.111% | 84.597% | 84.229% |
| 6 | ID34_UCF_BoyrazTappen | 82.829% | 82.640% | 83.352% | 82.496% |
| 7 | ID36_UCSD_MSRA_SJTU | 80.895% | 79.410% | 81.251% | 82.025% |
| 8 | ID28_USC | 77.360% | 76.154% | 77.704% | 78.222% |
| 9 | ID31_NII | 73.389% | 71.102% | 73.671% | 75.393% |
| 10 | ID44-UNITN | 70.504% | 70.446% | 69.797% | 71.270% |
| 11 | ID42-UEC | 66.261% | 65.157% | 66.726% | 66.899% |
| 12 | ID26_UMD | 65.948% | 65.218% | 65.385% | 67.240% |
| 13 | ID47_UNAL | 65.675% | 65.313% | 65.480% | 66.231% |
| 14 | ID32_Buffalo | 64.296% | 63.405% | 65.365% | 64.118% |
| 15 | ID29_TNO | 63.457% | 62.007% | 63.461% | 64.904% |
| 16 | ID37_ECNU | 54.738% | 54.764% | 55.162% | 54.287% |
Results of individual runs:
| Rank | Submission | Overall Accuracy | Split 1 Acc. | Split 2 Acc. | Split 3 Acc. |
| 1 | ID39_INRIA | 85.900% | 84.734% | 85.862% | 87.105% |
| 2 | ID40_Florence_Run4 | 85.708% | 85.319% | 86.642% | 85.164% |
| 3 | ID35_Canberra | 85.437% | 84.761% | 86.367% | 85.183% |
| 4 | ID40_Florence_Run3 | 84.442% | 83.703% | 85.556% | 84.066% |
| 5 | ID38_CAS_SIAT | 84.164% | 83.515% | 84.607% | 84.368% |
| 6 | ID25_Nanjing_Run2 | 83.979% | 83.111% | 84.597% | 84.229% |
| 7 | ID25_Nanjing_Run1 | 83.390% | 82.261% | 83.896% | 84.012% |
| 8 | ID34_UCF_BoyrazTappen_Run3 | 82.829% | 82.640% | 83.352% | 82.496% |
| 9 | ID40_Florence_Run2 | 82.455% | 81.468% | 83.013% | 82.883% |
| 10 | ID36_UCSD_MSRA_SJTU | 80.895% | 79.410% | 81.251% | 82.025% |
| 11 | ID34_UCF_BoyrazTappen_Run2 | 78.522% | 78.428% | 78.328% | 78.809% |
| 12 | ID34_UCF_BoyrazTappen_Run1 | 77.902% | 77.619% | 77.630% | 78.457% |
| 13 | ID28_USC | 77.360% | 76.154% | 77.704% | 78.222% |
| 14 | ID40_Florence_Run1 | 74.595% | 72.853% | 74.963% | 75.969% |
| 15 | ID31_NII | 73.389% | 71.102% | 73.671% | 75.393% |
| 16 | ID44_UNITN | 70.504% | 70.446% | 69.797% | 71.270% |
| 17 | ID42_UEC | 66.261% | 65.157% | 66.726% | 66.899% |
| 18 | ID26_UMD_Run2 | 65.948% | 65.218% | 65.385% | 67.240% |
| 19 | ID26_UMD_Run1 | 65.804% | 64.979% | 65.730% | 66.704% |
| 20 | ID47_UNAL_Run1 | 65.675% | 65.313% | 65.480% | 66.231% |
| 21 | ID32_Buffalo_Run2 | 64.296% | 63.405% | 65.365% | 64.118% |
| 22 | ID29_TNO | 63.457% | 62.007% | 63.461% | 64.904% |
| 23 | ID37_ECNU | 54.738% | 54.764% | 55.162% | 54.287% |
| 24 | ID32_Buffalo_Run1 | 45.606% | 44.677% | 45.480% | 46.662% |
| 25 | ID47_UNAL_Run2 | 1.109% | 1.219% | 0.891% | 1.217% |
**Detailed results and statistics available here:
PPT File.**
Notebook papers of the submissions:
Submission ID21_UOttawa:
LPM for Fast Action Recognition with Large Number of Classes,
Feng Shi, Robert Laganiere, Emil Petriu and Haiyu Zhen.
Confusion Tables: [
Run1 ,
Run2,
Run3 ,
Run4 ,
Run5]
Submission ID25_Nanjing:
Towards Good Practices for Action Video Encoding
, Jianxin Wu.
Confusion Tables: [
Run1 ,
Run2]
Submission ID26_UMD:
Evaluation of LC-KSVD on UCF101 Action Dataset ,
Hyunjong Cho, Hyungtae Lee, and Zhuolin Jiang.
Confusion Tables: [
Run1 ,
Run2]
Submission ID28_USC:
USC Action Recognition System with a Large Number of Classes,
Chen Sun and Ram Nevatia.
Confusion Table: [
Run1]
Submission ID29_TNO:
Action Recognition by Layout, Selective Sampling and Soft-Assignment
, G.J. Burghouts, P. Eendebak, H. Bouma and R.J-M. ten Hove.
Confusion Table: [
Run1]
Submission ID31_NII:
NII, Japan at the first THUMOSWorkshop 2013 ,
Sang Phan, Duy-Dinh Le and Shin'ichi Satoh.
Confusion Table: [
Run1]
Submission ID32_Buffalo:
Action Bank for Large-Scale Action Classification,
Wei Chen, Ran Xu, Jason J. Corso.
Confusion Tables: [
Run1 ,
Run2]
Submission ID34_UCF_BoyrazTappen:
Weakly-Supervised Action Recognition, Hakan Boyraz, Syed Masood, Baoyuan Liu, Marshall Tappen
Confusion Tables: [
Run1,
Run2,
Run3,
]
Submission ID35_Canberra:
Combined Ordered and Improved Trajectories for Large Scale Human Action Recognition,
O. V. Ramana Murthy and Roland Goecke.
Confusion Table: [
Run1]
Submission ID36_UCSD_MSRA_SJTU:
A Two-Layer Representation For Large-Scale Action Recognition,
Jun Zhu1, Baoyuan Wang, Xiaokang Yang, Wenjun Zhang, and Zhuowen Tu.
Confusion Table: [
Run1]
Submission ID37_ECNU:
Experimenting Motion Relativity for Action Recognition with a Large Number of Classes,
Feng Wang, Xiaoyan Li, and Wenmin Shu.
Confusion Table: [
Run1]
Submission ID38_CAS_SIAT:
Hybrid Super Vector with Improved Dense Trajectories for Action Recognition,
Xiaojiang Peng, LiMin Wang, Zhuowei Cai, Yu Qiao, and Qiang Peng.
Confusion Table: [
Run1]
Submission ID39_INRIA:
LEAR-INRIA submission for the THUMOS workshop,
Heng Wang and Cordelia Schmid.
Confusion Table: [
Run1]
Submission ID40_Florence:
L1-regularized Logistic Regression Stacking and Transductive CRF Smoothing for Action Recognition in Video,
Svebor Karaman, Lorenzo Seidenari, Andrew D. Bagdanov, Alberto Del Bimbo.
Confusion Tables: [
Run1,
Run2,
Run3,
Run4]
Submission ID42_UEC:
Fusion of Dense SURF Triangulation Features and Dense Trajectory based Features,
Do Hang Nga, Yoshiyuki Kawano, and Keiji Yanai.
Confusion Table: [
Run1]
Submission ID44_UNITN:
Action Recognition Using Accelerated Local Descriptors and Temporal Variation,
Negar Rostamzadeh, Jasper Uijlings and Nicu Sebe.
Confusion Table: [
Run1]
Submission ID47_UNAL:
MindLAB at the THUMOS Challeng,
Fabian Paez, Jorge A. Vanegas, and Fabio A. Gonzalez.
Confusion Tables: [
Run1 ,
Run2]
Submission ID47_UNAL:
MindLAB at the THUMOS Challeng,
Fabian Paez, Jorge A. Vanegas, and Fabio A. Gonzalez.
Confusion Tables: [
Run1 ,
Run2]
Submission ID20:
Ordered Trajectories for Large Scale Human Action Recognition,
O. V. Ramana Murthy, and Roland Goecke.
Submission ID22:
A Spatio-Temporal Feature based on Triangulation of Dense SURF,
Do Hang Nga, and Keiji Yanai.
Workshop Organization
General Chairs:
Ivan Laptev, INRIA
Massimo Piccardi, Univ. of Tech., Sydney
Mubarak Shah, UCF
Rahul Sukthankar, Google Research
Yu-Gang Jiang, Fudan University
Jingen Liu, SRI International
Amir Roshan Zamir, UCF
All names ordered alphabetically.
keynote speakers
Jason J. Corso, SUNY at Buffalo, USA
Tal Hassner, Open University, ISrael
Silvio Savarese, Stanford, USA
Cordelia Schmid, INRIA, France
Stan Sclaroff, Boston University, USA
Jianxin Wu, NJU, China
program committee
Saad Ali, SRI International, USA
Marco Bertini, University of Florence, Italy
Cigdem Beyan, University of Edinburgh, UK
Francois Bremond, INRIA, France
Liangliang Cao, IBM T. J. Watson, USA
Jason J. Corso, SUNY at Buffalo, USA
Riad Hammoud, BAE Systems, USA
nazli ikizler-cinbis, Hacettepe Univ., Turkey
Quoc V. Le, Stanford University, USA
Zicheng Liu, MSR, USA
Jiebo Luo, University of Rochester, USA
Greg Mori, SFU, Canada
Ronald Poppe, Univ. of Twente, Netherlands
Michalis Raptis, UCLA, USA
Michael S. Ryoo, NASA JPL, USA
Shin'ichi Satoh, NII, Japan
Silvio Savarese, Stanford, USA
Stan Sclaroff, Boston University, USA
Cees Snoek,Univ. of Amsterdam,Netherlands
Sinisa Todorovic, Oregon State Univ., USA
Yang Wang, University of Manitoba, Canada
Jianxin Wu, NJU, China
Lexing Xie, ANU, Australia
Shuicheng Yan, NUS, Singapore
Alper Yilmaz, OSU, USA
Junsong Yuan, NTU, Singapore
Lihi Zelnik-Manor, Technion, Israel
Data Collection
Khurram Soomro, UCF
sponsors
ICCV International Workshop on Action Recognition with a Large Number of Classes, Sydney, Australia, 2013