C
Cong Ma
Researcher at Peking University
Publications - 12
Citations - 249
Cong Ma is an academic researcher from Peking University. The author has contributed to research in topics: Video tracking & Deep learning. The author has an hindex of 5, co-authored 12 publications receiving 167 citations. Previous affiliations of Cong Ma include SenseTime.
Papers
More filters
Proceedings ArticleDOI
Trajectory Factory: Tracklet Cleaving and Re-Connection by Deep Siamese Bi-GRU for Multiple Object Tracking
TL;DR: In this article, a Siamese Bi-Gated Recurrent Unit (GRU) based tracklet re-connection method is applied to link the sub-tracklets which belong to the same object to form a whole trajectory.
Proceedings ArticleDOI
Dense Relation Network: Learning Consistent and Context-Aware Representation for Semantic Image Segmentation
TL;DR: This paper proposes dense relation network (DRN) and context-restricted loss (CRL) to aggregate global and local information to make the best of global context.
Proceedings ArticleDOI
Deep Association: End-to-end Graph-Based Learning for Multiple Object Tracking with Conv-Graph Neural Network
TL;DR: An efficient end-to-end model, Deep Association Network (DAN), to learn the graph-based training data, which are constructed by spatial-temporal interaction of objects, up to the state-of-the-art methods without extra-dataset on MOT16 and DukeMTMCT.
Posted Content
Trajectory Factory: Tracklet Cleaving and Re-connection by Deep Siamese Bi-GRU for Multiple Object Tracking
TL;DR: A novel tracklet processing method to cleave and re-connect tracklets on crowd or longterm occlusion by Siamese Bi-Gated Recurrent Unit (GRU) to create the high-confidence tracklet candidates in sparse scenario.
Proceedings ArticleDOI
RelationNet: Learning Deep-Aligned Representation for Semantic Image Segmentation
TL;DR: A novel deep neural network named Relation net is proposed, which utilizes CNN and RNN to aggregate context information and a spatial correlation loss is applied to train RelationNet to align features of spatial pixels belonging to same category.