J
Jianchao Yang
Researcher at Adobe Systems
Publications - 184
Citations - 28091
Jianchao Yang is an academic researcher from Adobe Systems. The author has contributed to research in topics: Convolutional neural network & Sparse approximation. The author has an hindex of 60, co-authored 183 publications receiving 24321 citations. Previous affiliations of Jianchao Yang include University of Illinois at Urbana–Champaign.
Papers
More filters
Patent
Image classification using images with separate grayscale and color channels
TL;DR: In this article, an image classification network includes grayscale filters and color filters, which are separate from the grayscale filters, and are used to identify an object in the image, and the image is classified based on the identified object.
Proceedings Article
Pooling robust shift-invariant sparse representations of acoustic signals
TL;DR: This paper investigates different pooling strategies based on the sparse coding scheme and proposes a temporal pyramid pooling method to extract discriminative and shiftinvariant feature representations and demonstrates the superiority of the new feature representation over traditional features on the acoustic event classification task.
Proceedings ArticleDOI
Action Recognition with Visual Attention on Skeleton Images
TL;DR: This work redesigns skeleton representations with a depth-first tree traversal order, which enhances the semantic meaning of skeleton images and better preserves the structural information, and proposes a two-branch attention architecture that focuses on spatio-temporal key stages and filters out unreliable joint predictions.
Patent
Finding semantic parts in images
TL;DR: In this paper, a convolutional neural network (CNN) is applied to a set of images to extract features for each image, and each feature is defined by a feature vector that enables a subset of the images to be clustered in accordance with a similarity between feature vectors.
Journal ArticleDOI
Image-Specific Prior Adaptation for Denoising
TL;DR: This work proposes a novel prior learning algorithm that combines the strength of both internal and external priors, and first learns a generic Gaussian mixture model from a collection of training images and adapt the model to the given image by simultaneously adding additional components and refining the component parameters.