PPaperPicks

Ming Lu

Intel Lab China, Cognitive Computing Laboratory, Beijing, China

25 papers at tracked venues · 20 at CORE A* · active 20242026

Venues

Frequent coauthors

Papers

  1. FastDriveVLA: Efficient End-to-End Driving via Plug-and-Play Reconstruction-based Token Pruning
  2. MMG-Vid: Maximizing Marginal Gains at Segment-level and Token-level for Efficient Video LLMs
  3. ManipDreamer3D: Synthesizing Plausible Robotic Manipulation Video with Occupancy-aware 3D Trajectory
  4. StreamKV: Streaming Video Question-Answering with Segment-based KV Cache Retrieval and Compression
  5. 3DRealCar: An In-the-Wild RGB-D Car Dataset with 360-Degree Views
  6. Beyond Attention or Similarity: Maximizing Conditional Diversity for Token Pruning in MLLMs
  7. Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMs
  8. DepthDark: Robust Monocular Depth Estimation for Low-Light Environments
  9. EMD: Explicit Motion Modeling for High-Quality Street Gaussian Splatting
  10. EmbodiedOcc++: Boosting Embodied 3D Occupancy Prediction with Plane Regularization and Uncertainty Sampler
  11. GazeGaussian: High-Fidelity Gaze Redirection with 3D Gaussian Splatting
  12. GraphAvatar: Compact Head Avatars with GNN-Generated 3D Gaussians
  13. K-Buffers: A Plug-in Method for Enhancing Neural Fields with Multiple Buffers
  14. MixedGaussianAvatar: Realistically and Geometrically Accurate Head Avatar via Mixed 2D-3D Gaussians
  15. MoVE-KD: Knowledge Distillation for VLMs with Mixture of Visual Encoders
  16. SliceOcc: Indoor 3D Semantic Occupancy Prediction with Vertical Slice Representation
  17. ThermalGaussian: Thermal 3D Gaussian Splatting
  18. UniCTokens: Boosting Personalized Understanding and Generation via Unified Concept Tokens
  19. VGNC: Reducing the Overfitting of Sparse-view 3DGS via Validation-guided Gaussian Number Control
  20. BEVUDA: Multi-geometric Space Alignments for Domain Adaptive BEV 3D Object Detection
  21. I-MedSAM: Implicit Medical Image Segmentation with Segment Anything
  22. NTO3D: Neural Target Object 3D Reconstruction with Segment Anything
  23. Superpixel-based Efficient Sampling for Learning Neural Fields from Large Input
  24. Unsupervised Spike Depth Estimation via Cross-modality Cross-domain Knowledge Transfer
  25. ViDA: Homeostatic Visual Domain Adapter for Continual Test Time Adaptation