• Publications
  • Influence
Filter Pruning via Geometric Median for Deep Convolutional Neural Networks Acceleration
TLDR
Unlike previous methods, FPGM compresses CNN models by pruning filters with redundancy, rather than those with“relatively less” importance, and when applied to two image classification benchmarks, the method validates its usefulness and strengths. Expand
Person Re-identification in the Wild
TLDR
A new dataset, PRW, is introduced to evaluate Person Re-identification in the Wild, and it is shown that pedestrian detection aids re-ID through two simple yet effective improvements: a cascaded fine-tuning strategy that trains a detection model first and then the classification model, and a Confidence Weighted Similarity (CWS) metric that incorporates detection scores into similarity measurement. Expand
Improving Person Re-identification by Attribute and Identity Learning
TLDR
An attribute-person recognition (APR) network is proposed, a multi-task network which learns a re-ID embedding and at the same time predicts pedestrian attributes, and demonstrates that by learning a more discriminative representation, APR achieves competitive re-IDs performance compared with the state-of-the-art methods. Expand
Learning to Propagate Labels: Transductive Propagation Network for Few-Shot Learning
TLDR
This paper proposes Transductive Propagation Network (TPN), a novel meta-learning framework for transductive inference that classifies the entire test set at once to alleviate the low-data problem. Expand
High lithium anodic performance of highly nitrogen-doped porous carbon prepared from a metal-organic framework.
Theoretical and experimental results have revealed that the lithium-ion storage capacity for nitrogen-doped graphene largely depends on the nitrogen-doping level. However, most nitrogen-doped carbonExpand
Influence of zeolite crystal size on zeolite-polyamide thin film nanocomposite membranes.
  • M. Lind, A. Ghosh, +4 authors E. Hoek
  • Materials Science, Medicine
  • Langmuir : the ACS journal of surfaces and…
  • 15 June 2009
TLDR
The data presented offer additional support for the hypothesis that zeolite crystals alter polyamide thin film structure when they are present during the interfacial polymerization reaction. Expand
Dynamic Computational Time for Visual Attention
TLDR
A dynamic computational time model to accelerate the average processing time for recurrent visual attention (RAM) by adding an additional continue/stop action per time step to RAM and using reinforcement learning to learn both the optimal attention policy and stopping policy. Expand
Surface charge and cellular processing of covalently functionalized multiwall carbon nanotubes determine pulmonary toxicity.
  • Ruibin Li, Xiang Wang, +15 authors T. Xia
  • Materials Science, Medicine
  • ACS nano
  • 28 February 2013
TLDR
It is demonstrated that surface charge plays an important role in the structure-activity relationships that determine the pro-fibrogenic potential of f-CNTs in the lung. Expand
Dispersal state of multiwalled carbon nanotubes elicits profibrogenic cellular responses that correlate with fibrogenesis biomarkers and fibrosis in the murine lung.
  • Xiang Wang, T. Xia, +13 authors A. Nel
  • Materials Science, Medicine
  • ACS nano
  • 27 December 2011
TLDR
The results indicate that the dispersal state of MWCNTs affects profibrogenic cellular responses that correlate with the extent of pulmonary fibrosis and are of potential use to predict pulmonary toxicity. Expand
Dual plasmonic nanostructures for high performance inverted organic solar cells.
Polymer-fullerene-based bulk heterojunction (BHJ) solar cells have many advantages, including low-cost, low-temperature fabrication, semi-transparency, and mechanical fl exibility. [ 1 , 2 ] However,Expand
...
1
2
3
4
5
...