Collaborative Global-Local Structure Network with Knowledge Distillation for Imbalanced Data Classification

F Wu, Z Liu, Z Zhang, J Liu… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Multi-expert networks have shown great superiority for imbalanced data classification tasks
due to their complementary and diverse. We have summarized two aspects for further …

Highly realistic synthetic dataset for pixel-level densepose estimation via diffusion model

J Wen, T Chu, Q Liu - Pattern Recognition, 2024 - Elsevier
Generating training data with pixel-level annotations for DensePose is a labor-intensive
task, resulting in sparse labeling in real-world datasets. Prior solutions have relied on …

Dynamic Learnable Logit Adjustment for Long-Tailed Visual Recognition

E Zhang, C Geng, C Li, S Chen - IEEE Transactions on Circuits …, 2024 - ieeexplore.ieee.org
Logit adjustment is an effective long-tailed visual recognition strategy to encourage a
significant margin between rare and dominant labels. Existing methods typically employ the …

Devil in the details: Delving into accurate quality scoring for DensePose

J Sun, Q Liu - Pattern Recognition, 2024 - Elsevier
How to score the quality of the network output is an essential but long-neglected problem in
DensePose, which dramatically limits the potential of the existing methods. To fill the blank …

CoNPL: Consistency training framework with noise-aware pseudo labeling for dense pose estimation

J Wen, T Chu, J Sun, Q Liu - Image and Vision Computing, 2024 - Elsevier
Dense pose estimation faces hurdles due to the lack of costly precise pixel-level IUV labels.
Existing methods aim to overcome it by regularizing model outputs or interpolating pseudo …

GC RCNN: Learning Interdependent IUV for Geodesic Consistent DensePose

J Sun, Q Liu - IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS …, 2023 - techrxiv.org
The inconsistency between training strategies and task objectives is a principal constraint
impeding DensePose development. DensePose aims to minimize the surface distance …