E
Eric Chang
Researcher at Microsoft
Publications - 142
Citations - 7771
Eric Chang is an academic researcher from Microsoft. The author has contributed to research in topics: Image segmentation & Deep learning. The author has an hindex of 47, co-authored 138 publications receiving 6487 citations. Previous affiliations of Eric Chang include Massachusetts Institute of Technology & University of California, Irvine.
Papers
More filters
Proceedings ArticleDOI
Deep learning of feature representation with multiple instance learning for medical image analysis
TL;DR: In this article, the authors used multiple instance learning (MIL) framework in classification training with deep learning features and found that automatic feature learning outperformed manual feature learning and achieved performance that's close to fully supervised approach.
Proceedings ArticleDOI
Forecasting Fine-Grained Air Quality Based on Big Data
TL;DR: This paper forecasts the reading of an air quality monitoring station over the next 48 hours, using a data-driven method that considers current meteorological data, weather forecasts, and air quality data of the station and that of other stations within a few hundred kilometers.
Journal ArticleDOI
Large scale tissue histopathology image classification, segmentation, and visualization via deep convolutional activation features.
Yan Xu,Yan Xu,Zhipeng Jia,Zhipeng Jia,Liang-Bo Wang,Liang-Bo Wang,Yuqing Ai,Yuqing Ai,Fang Zhang,Fang Zhang,Maode Lai,Eric Chang +11 more
TL;DR: The framework proposed is a simple, efficient and effective system for histopathology image automatic analysis that successfully transfer ImageNet knowledge as deep convolutional activation features to the classification and segmentation of histopathological images with little training data.
Journal ArticleDOI
Weakly supervised histopathology cancer image segmentation and classification
TL;DR: This paper embeds the clustering concept into the multiple instance learning (MIL) setting and derive a principled solution to performing the above three tasks in an integrated framework and introduces contextual constraints as a prior for MCIL, which further reduces the ambiguity in MIL.
Proceedings ArticleDOI
Inferring gas consumption and pollution emission of vehicles throughout a city
TL;DR: The method instantly infers the gas consumption and pollution emission of vehicles traveling on a city's road network in a current time slot, using GPS trajectories from a sample of vehicles (e.g., taxicabs) based on a context-aware matrix factorization approach.