Kecheng Zheng

26

Papers

324

Total Citations

Papers (26)

Paying More Attention to Images: A Training-Free Method for Alleviating Hallucination in LVLMs

Language-Image Pre-training with Long Captions

Animate-X: Universal Character Image Animation with Enhanced Motion Representation

Benchmarking Large Vision-Language Models via Directed Scene Graph for Comprehensive Image Captioning

Mimir: Improving Video Diffusion Models for Precise Text Understanding

Exploring Sparse MoE in GANs for Text-conditioned Image Synthesis

MotionStone: Decoupled Motion Intensity Modulation with Diffusion Transformer for Image-to-Video Generation

Aligned Better, Listen Better for Audio-Visual Large Language Models

Contextual AD Narration with Interleaved Multimodal Sequence

Learning Visual Generative Priors without Text

Self-Organizing Pathway Expansion for Non-Exemplar Class-Incremental Learning

Spatial-Temporal Correlation and Topology Learning for Person Re-Identification in Videos

CoDeF: Content Deformation Fields for Temporally Consistent Video Processing

CrossMAE: Cross-Modality Masked Autoencoders for Region-Aware Audio-Visual Pre-Training

Group-aware Label Transfer for Domain Adaptive Person Re-identification

Unleashing Potential of Unsupervised Pre-Training With Intra-Identity Regularization for Person Re-Identification

Cloth-Changing Person Re-Identification From a Single Image With Gait Prediction and Regularization

Temporal Complementarity-Guided Reinforcement Learning for Image-to-Video Person Re-Identification

Neural Dependencies Emerging From Learning Massive Categories

Uncertainty-Aware Optimal Transport for Semantically Coherent Out-of-Distribution Detection

Regularized Mask Tuning: Uncovering Hidden Knowledge in Pre-Trained Vision-Language Models

Abstract Reasoning with Distracting Features

Uncertainty-Aware Hierarchical Refinement for Incremental Implicitly-Refined Classification

Rank Diminishing in Deep Neural Networks

Benchmarking and Analyzing 3D-aware Image Synthesis with a Modularized Codebase

Customizable Image Synthesis with Multiple Subjects