Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation
- C. Li, Qiaoyong Zhong, Di Xie, Shiliang Pu
- Computer ScienceInternational Joint Conference on Artificial…
- 17 April 2018
This paper introduces a global spatial aggregation scheme, which is able to learn superior joint co-occurrence features over local aggregation, and consistently outperforms other state-of-the-arts on action recognition and detection benchmarks like NTU RGB+D, SBU Kinect Interaction and PKU-MMD.
Multi-Interest Network with Dynamic Routing for Recommendation at Tmall
- C. Li, Zhiyuan Liu, Lee
- Computer ScienceInternational Conference on Information and…
- 17 April 2019
This paper designs a multi-interest extractor layer based on the recently proposed dynamic routing mechanism, which is applicable for modeling and extracting diverse interests from user's behaviors, and proposes a technique named label-aware attention to help the learning process of user representations.
Robust depth estimation for light field via spinning parallelogram operator
- Shuo Zhang, Hao Sheng, C. Li, Jun Zhang, Z. Xiong
- Computer ScienceComputer Vision and Image Understanding
- 1 April 2016
Expanded base editing in rice and wheat using a Cas9-adenosine deaminase fusion
- C. Li, Y. Zong, Caixia Gao
- BiologyGenome Biology
- 29 May 2018
A new plant adenine base editor based on an evolved tRNA adenosine deaminase fused to the nickase CRISPR/Cas9 is described, enabling A•T to G•C conversion at frequencies up to 7.5% in protoplasts and 59.1% in regenerated rice and wheat plants.
Skeleton-based action recognition with convolutional neural networks
- C. Li, Qiaoyong Zhong, Di Xie, Shiliang Pu
- Computer ScienceIEEE International Conference on Multimedia…
- 25 April 2017
A novel convolutional neural networks (CNN) based framework for both action classification and detection of skeleton-based action recognition and a window proposal network to extract temporal segment proposals, which are further classified within the same network.
yaSpMV: yet another SpMV framework on GPUs
- Shengen Yan, C. Li, Yunquan Zhang, Huiyang Zhou
- Computer ScienceACM SIGPLAN Symposium on Principles & Practice of…
- 6 February 2014
A new SpMV format is devised, called blocked compressed common coordinate (BCCOO), which uses bit flags to store the row indices in a blocked common coordinate format so as to alleviate the bandwidth problem and an auto-tuning framework is introduced to choose optimization parameters based on the characteristics of input sparse matrices and target hardware platforms.
Non-Local Meets Global: An Integrated Paradigm for Hyperspectral Denoising
- Wei He, Quanming Yao, C. Li, N. Yokoya, Qibin Zhao
- Computer ScienceComputer Vision and Pattern Recognition
- 11 December 2018
A unified spatial-spectral paradigm for HSI denoising is proposed by first learning a low-dimensional orthogonal basis and the related reduced image from the noisy HSI, and an efficient algorithm motivated by alternating minimization is developed.
Tensor Ring Decomposition with Rank Minimization on Latent Space: An Efficient Approach for Tensor Completion
- Longhao Yuan, C. Li, D. Mandic, Jianting Cao, Qibin Zhao
- Computer ScienceAAAI Conference on Artificial Intelligence
- 7 September 2018
This paper proposes a novel tensor completion method which is robust to model selection, and introduces nuclear norm regularization on the latent TR factors, resulting in the optimization step using singular value decomposition (SVD) being performed at a much smaller scale.
Efficient C-to-T base editing in plants using a fusion of nCas9 and human APOBEC3A
- Y. Zong, Qianna Song, Caixia Gao
- BiologyNature Biotechnology
- 1 October 2018
It is shown that a base editor fusion protein composed of Cas9 nickase and human APOBEC3A (A3A-PBE) converts cytidine to thymidine efficiently in wheat, rice and potato with a 17-nucleotide editing window at all examined sites, independent of sequence context.
...
...