3 papers
cs.CV2024
Incorporating Test-Time Optimization into Training with Dual Networks for Human Mesh Recovery
Yongwei Nie, Mingxian Fan, Chengjiang Long +3
Human Mesh Recovery (HMR) is the task of estimating a parameterized 3D human mesh from an image. There is a kind of methods first training a regression model for this problem, then…
cs.CV2024
Multi-RoI Human Mesh Recovery with Camera Consistency and Contrastive Losses
Yongwei Nie, Changzhen Liu, Chengjiang Long +3
Besides a 3D mesh, Human Mesh Recovery (HMR) methods usually need to estimate a camera for computing 2D reprojection loss. Previous approaches may encounter the following problem:…
cs.CV2024
Interleaving One-Class and Weakly-Supervised Models with Adaptive Thresholding for Unsupervised Video Anomaly Detection
Yongwei Nie, Hao Huang, Chengjiang Long +3
Video Anomaly Detection (VAD) has been extensively studied under the settings of One-Class Classification (OCC) and Weakly-Supervised learning (WS), which however both require labo…