Cheng Jin

11

Papers

62

Total Citations

Papers (11)

High-fidelity Person-centric Subject-to-Image Synthesis

FusionFormer: A Concise Unified Feature Fusion Transformer for 3D Pose Estimation

DreamText: High Fidelity Scene Text Synthesis

Towards Robust Influence Functions with Flat Validation Minima

Population Normalization for Federated Learning

Point Cloud Part Editing: Segmentation, Generation, Assembly, and Selection

PoseIRM: Enhance 3D Human Pose Estimation on Unseen Camera Settings via Invariant Risk Minimization

An Exemplar-based Framework for Chinese Text Recognition

Optimized Gradient Clipping for Noisy Label Learning

Achieving Ensemble-Like Performance in a Single Model: A Feature Diversification Framework for Image-Text Matching

Expanding the Scope of Negatives: Boosting Image-Text Matching with Negatives Distribution Guided Learning