S
Song Wang
Researcher at University of South Carolina
Publications - 272
Citations - 7824
Song Wang is an academic researcher from University of South Carolina. The author has contributed to research in topics: Image segmentation & Computer science. The author has an hindex of 34, co-authored 231 publications receiving 5276 citations. Previous affiliations of Song Wang include University of Illinois at Urbana–Champaign & Tianjin University.
Papers
More filters
Proceedings ArticleDOI
Learning Dynamic Siamese Network for Visual Object Tracking
TL;DR: This paper proposes dynamic Siamese network, via a fast transformation learning model that enables effective online learning of target appearance variation and background suppression from previous frames, and presents elementwise multi-layer fusion to adaptively integrate the network outputs using multi-level deep features.
Journal ArticleDOI
CrackTree: Automatic crack detection from pavement images
TL;DR: The proposed CrackTree method is evaluated on a collection of 206 real pavement images and the experimental results show that the proposed method achieves a better performance than several existing methods.
Journal ArticleDOI
DeepCrack: Learning Hierarchical Convolutional Features for Crack Detection
TL;DR: DeepCrack-an end-to-end trainable deep convolutional neural network for automatic crack detection by learning high-level features for crack representation and outperforms the current state-of-the-art methods.
Journal ArticleDOI
Image segmentation with ratio cut
Song Wang,Jeffrey Mark Siskind +1 more
TL;DR: A new cost function, cut ratio, for segmenting images using graph-based methods that allows the image perimeter to be segmented, guarantees that the segments produced by bipartitioning are connected, and does not introduce a size, shape, smoothness, or boundary-length bias.
Proceedings ArticleDOI
Recognize Human Activities from Partially Observed Videos
Yu Cao,Daniel Paul Barrett,Andrei Barbu,Siddharth Narayanaswamy,Haonan Yu,Aaron Michaux,Yuewei Lin,Sven Dickinson,Jeffrey Mark Siskind,Song Wang +9 more
TL;DR: A new method that can recognize human activities from partially observed videos in the general case by dividing each activity into multiple ordered temporal segments and applying sparse coding to derive the activity likelihood of the test video sample at each segment is proposed.