• Publications
  • Influence
Feature Mining for Localised Crowd Counting
TLDR
This paper presents a single regression model based approach that is able to estimate people count in spatially localised regions and is more scalable without the need for training a large number of regressors proportional to the number of local regions. Expand
CACTI-3DD: Architecture-level modeling for 3D die-stacked DRAM main memory
TLDR
CACTI-3DD is introduced, the first architecture-level integrated power, area, and timing modeling framework for 3D die-stacked off-chip DRAM main memory, and the results show that the 3D DRAM with re-architected DRAM dies achieves significant improvements in power and timing compared to the coarse-grained 3DDie-Stacked DRAM. Expand
Cumulative Attribute Space for Age and Crowd Density Estimation
TLDR
This paper introduces a novel cumulative attribute concept for learning a regression model when only sparse and imbalanced data are available, and gains notable advantage on accuracy for both age estimation and crowd counting when compared against conventional regression models. Expand
CACTI-P: Architecture-level modeling for SRAM-based structures with advanced leakage reduction techniques
TLDR
It is found that although nanosecond scale power-gating is a powerful way to minimize leakage power for all levels of caches, its severe impacts on processor performance and energy when being used for L1 data caches make nanose Cond scalePower-Gating a better fit for caches closer to main memory. Expand
Gerontechnology acceptance by elderly Hong Kong Chinese: a senior technology acceptance model (STAM)
TLDR
By encompassing conventional technology acceptance constructs together with age-related health and ability characteristics, the present study was able to identify more factors affecting gerontechnology acceptance by older Hong Kong Chinese. Expand
Structured Knowledge Distillation for Semantic Segmentation
TLDR
This paper starts from the straightforward scheme, pixel-wise distillation, which applies the distillation scheme originally introduced for image classification and performs knowledge distillation for each pixel separately, and proposes to distill the structured knowledge from cumbersome networks into compact networks, motivated by the fact that semantic segmentation is a structured prediction problem. Expand
Semi-Supervised Learning via Regularized Boosting Working on Multiple Semi-Supervised Assumptions
  • Ke Chen, Shihai Wang
  • Mathematics, Computer Science
  • IEEE Transactions on Pattern Analysis and Machine…
  • 2011
TLDR
This paper proposes a novel cost functional consisting of the margin cost on labeled data and the regularization penalty on unlabeled data based on three fundamental semi-supervised assumptions and demonstrates that the algorithm yields favorite results for benchmark and real-world classification tasks in comparison to state-of-the-art semi- supervised learning algorithms, including newly developed boosting algorithms. Expand
Induction of leptin resistance through direct interaction of C-reactive protein with leptin
TLDR
The presence in human blood of several serum leptin-interacting proteins (SLIPs), isolated by leptin-affinity chromatography and identified by mass spectrometry and immunochemical analysis, confirmed that one of the major SLIPs is C-reactive protein (CRP). Expand
Sumblr: continuous summarization of evolving tweet streams
TLDR
This paper proposes a novel prototype called Sumblr (SUMmarization By stream cLusteRing) for tweet streams, and develops a TCV-Rank summarization technique for generating online summaries and historical summaries of arbitrary time durations. Expand
Image selective segmentation under geometrical constraints using an active contour approach
TLDR
A new model for segmentation of an image under some geometrical constraints in order to detect special regions of interest is proposed by combining it with the idea of a piecewise constant Mumford-Shah model as with the non-selective Chan-Vese segmentation. Expand
...
1
2
3
4
5
...