Y
Yan Song
Researcher at Nanjing University of Science and Technology
Publications - 43
Citations - 730
Yan Song is an academic researcher from Nanjing University of Science and Technology. The author has contributed to research in topics: Sparse approximation & Feature extraction. The author has an hindex of 14, co-authored 43 publications receiving 652 citations. Previous affiliations of Yan Song include University of California, San Diego & Chinese Academy of Sciences.
Papers
More filters
Proceedings ArticleDOI
Inception Single Shot MultiBox Detector for object detection
TL;DR: This paper adopts the Inception block to replace the extra layers in SSD, and calls this method Inception SSD (I-SSD), and proposes an improved non-maximum suppression method to overcome its deficiency on the expression ability of the model.
Proceedings ArticleDOI
Semi-automatic video annotation based on active learning with multiple complementary predictors
TL;DR: It is proved that the samples selected by the proposed scheme are more representative than general active learning scheme, as well as the incremental model adaptation scheme is effective especially when the newly added data size is small.
Journal ArticleDOI
Body Surface Context: A New Robust Feature for Action Recognition From Depth Videos
TL;DR: This work proposes a new robust feature, the body surface context (BSC), by describing the distribution of relative locations of the neighbors for a reference point in the point cloud in a compact and descriptive way and proposes three schemes to represent human actions based on the new feature.
Proceedings ArticleDOI
A Novel Image Text Extraction Method Based on K-Means Clustering
TL;DR: A coarse-to-fine text location method is implemented, a multi-scale approach is adopted to locate texts with different font sizes, and color-based k-means clustering is adopted in text segmentation.
Journal ArticleDOI
Video Annotation Based on Kernel Linear Neighborhood Propagation
TL;DR: KLNP improves a recently proposed method linear neighborhood propagation by tackling the limitation of its local linear assumption on the distribution of semantics by combining the consistency assumption and the local linear embedding method in a nonlinear kernel-mapped space.