Conference

Multi-region two-stream R-CNN for action detection
Xiaojiang Peng, Cordelia Schmid
European Conference on Computer Vision (ECCV), 2016. [ New ]
[ PDF ] [ Bibtex ] [ Poster] [Code]*The results on UCF101 are updated.
Obtain the state-of-the-art results on UCF-Sports, JHMDB, and UCF101 datasets.

Encoding Feature Maps of CNNs for Action Recognition
Xiaojiang Peng, Cordelia Schmid
THUMOS Challenge 2015 Workshop, CVPR, 2015.
[ PDF ] [ Bibtex ] [ Slides ]

Action Recognition with Stacked Fisher Vectors
Xiaojiang Peng, C.Q. Zou, Y. Qiao, and Q. Peng
European Conference on Computer Vision (ECCV), Zurich, Switzerland, 2014
[ PDF ] [ Bibtex ] [ Poster ] [ Code& 2nd FVs pw: 5480 or here ]

Boosting VLAD with Supervised Dictionary Learning and High-Order Statistics
Xiaojiang Peng, L.M. Wang, Y. Qiao, and Q. Peng
European Conference on Computer Vision (ECCV), Zurich, Switzerland, 2014
[ PDF ] [ Bibtex ] [ Poster ]

Action and Gesture Temporal Spotting with Super Vector Representation 
Xiaojiang Peng, L.M. Wang, Z.W. Cai, and Y. Qiao 
ChaLearn Looking at People (LAP) workshop, ECCV, 2014.
[ PDF ] [ Bibtex ] 
 Winner of both track 1 and track2, and rank 4th for track3.

Multi-View Super Vector for Action Recognition 
Z.W, Cai, L.M. Wang, Xiaojiang Peng, and Y. Qiao. 
IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), Columbus, Ohio, USA, June 2014. (oral) 
[ PDF ] [ Bibtex ] [ Video Spotlight ]

A Joint Evaluation of Dictionary Learning and Feature Encoding for Action Recognition 
Xiaojiang Peng, L.M. Wang, Y. Qiao and Q. Peng 
International Conference on Pattern Recognition (ICPR), Stockholm, Sweden, 2014
[ PDF ] [ Bibtex ] [ Poster ]

Exploring Motion Boundary Based Sampling and Spatial-temporal Context Descriptors for Action Recognition 
Xiaojiang Peng, Y. Qiao, Q. Peng and X.B. Qi 
British Machine Vision Conference (BMVC), Bristol, United Kingdom, 2013.
[ PDF ] [ Bibtex ] [ Poster ] [ Code ]

Hybrid Super Vector with Improved Dense Trajectories for Action Recognition 
Xiaojiang Peng, L.M. Wang, Z.W. Cai, Y. Qiao and Q. Peng 
THUMOS'13 Action Recognition Challenge Workshop, ICCV, 2013. 
[ PDF ] [ Bibtex ] [ Slides ]

Journal

Bag of Visual Words and Fusion Methods for Action Recognition: Comprehensive Study and Good Practice 
X.J. Peng, L. Wang, X.X. Wang, and Y. Qiao 
Computer Vision and Image Understanding, 2016  
[ PDF ]  [ Code ]

Large Margin Dimensionality Reduction for Action Similarity Labeling 
Xiaojiang Peng, Y. Qiao, Q. Peng and Q.H. Wang 
IEEE Signal Processing Letter, 2014.
[ PDF ] [ Bibtex ] [ Code ]

Motion Boundary Based Sampling and 3D Co-occurrence Descriptors for Action Recognition
Xiaojiang Peng, Y. Qiao, Q. Peng 
Image and Vision Computing, 2014. 
[ PDF ] [ Bibtex ] [ Code ]

Technical reports

Frankenstein: Learning Deep Face Representations using Small Data
Guosheng Hu, Xiaojiang Peng, Yongxin Yang, Timothy Hospedales, Jakob Verbeek 
arXiv:1603.06470 , 2016.  [ New ]
[ PDF ]