Ming Lu

12

Papers

63

Total Citations

Papers (12)

Beyond Attention or Similarity: Maximizing Conditional Diversity for Token Pruning in MLLMs

3DRealCar: An In-the-wild RGB-D Car Dataset with 360-degree Views

Another Way to the Top: Exploit Contextual Clustering in Learned Image Coding

GraphAvatar: Compact Head Avatars with GNN-Generated 3D Gaussians

Towards Backward-Compatible Continual Learning of Image Compression

GazeGaussian: High-Fidelity Gaze Redirection with 3D Gaussian Splatting

EMD: Explicit Motion Modeling for High-Quality Street Gaussian Splatting

Towards Loss-Resilient Image Coding for Unstable Satellite Networks

Deep Hierarchical Video Compression

Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMs

NTO3D: Neural Target Object 3D Reconstruction with Segment Anything

MoVE-KD: Knowledge Distillation for VLMs with Mixture of Visual Encoders