P
Peng Tang
Researcher at Huazhong University of Science and Technology
Publications - 41
Citations - 2521
Peng Tang is an academic researcher from Huazhong University of Science and Technology. The author has contributed to research in topics: Convolutional neural network & Object detection. The author has an hindex of 19, co-authored 37 publications receiving 1648 citations. Previous affiliations of Peng Tang include Salesforce.com & Johns Hopkins University.
Papers
More filters
Proceedings ArticleDOI
Multiple Instance Detection Network with Online Instance Classifier Refinement
TL;DR: This work formulate weakly supervised object detection as a Multiple Instance Learning (MIL) problem, where instance classifiers (object detectors) are put into the network as hidden nodes and instance labels inferred from weak supervision are propagated to their spatially overlapped instances to refine instance classifier online.
Journal ArticleDOI
Revisiting multiple instance neural networks
TL;DR: This article revisit Multiple Instance Neural Networks (MINNs) that the neural networks aim at solving the MIL problems and proposes a new type of MINN to learn bag representations, which is different from the existing MINNs that focus on estimating instance label.
Journal ArticleDOI
PCL: Proposal Cluster Learning for Weakly Supervised Object Detection
TL;DR: In this article, the authors propose to use proposal clusters to learn refined instance classifiers by an iterative process, where the proposals in the same cluster are spatially adjacent and associated with the same object.
Book ChapterDOI
Weakly Supervised Region Proposal Network and Object Detection
Peng Tang,Xinggang Wang,Angtian Wang,Yongluan Yan,Wenyu Liu,Junzhou Huang,Junzhou Huang,Alan L. Yuille +7 more
TL;DR: This paper proposes a weakly supervised region proposal network which is trained using only image-level annotations and achieves the state-of-the-art performance for WSOD with performance gain of about \(3\%\) on average.
Posted Content
PCL: Proposal Cluster Learning for Weakly Supervised Object Detection
TL;DR: This paper first shows that instances can be assigned object or background labels directly based on proposal clusters for instance classifier refinement, and then shows that treating each cluster as a small new bag yields fewer ambiguities than the directly assigning label method.