C
Chuanqi Dong
Researcher at Nanjing University
Publications - 9
Citations - 58
Chuanqi Dong is an academic researcher from Nanjing University. The author has contributed to research in topics: Computer science & Image warping. The author has an hindex of 2, co-authored 6 publications receiving 13 citations.
Papers
More filters
Proceedings ArticleDOI
Learning Task-aware Local Representations for Few-shot Learning.
TL;DR: An Adaptive Task-aware Local Representations Network (ATL-Net) is proposed to address this limitation by introducing episodic attention, which can adaptively select the important local patches among the entire task, as the process of human recognition.
Proceedings ArticleDOI
Biased Feature Learning for Occlusion Invariant Face Recognition
TL;DR: A Biased Feature Learning framework for occlusion-invariant face recognition that enhances the robustness of a network to unknown occlusions but also maintains or even improves its performance for normal faces.
Journal ArticleDOI
CariMe: Unpaired Caricature Generation with Multiple Exaggerations
TL;DR: A Multi-exaggeration Warper network is proposed to learn the distribution-level mapping from photos to facial exaggerations, which makes it possible to generate diverse and reasonable exaggerations from randomly sampled warp codes given one input photo.
Proceedings ArticleDOI
DeepMEF: A Deep Model Ensemble Framework for Video Based Multi-modal Person Identification
TL;DR: The proposed deep model ensemble framework includes three novel modules, namely DeepMEF, which adopts the scene feature extracted by ourselves as the additional input of the multi-modal module and promotes the overall performance by combining the predictions of multiple multi- modal learners.
Posted Content
CariMe: Unpaired Caricature Generation with Multiple Exaggerations
TL;DR: CariMe as discussed by the authors generalizes the caricature generation problem from instance-level warping prediction to distribution-level deformation modeling, which makes it possible to generate diverse and reasonable exaggerations from randomly sampled warp codes given one input photo.