Wenzhao Zheng

29

Papers

243

Total Citations

Papers (29)

SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference

DeSiRe-GS: 4D Street Gaussians for Static-Dynamic Decomposition and Surface Reconstruction for Urban Driving Scenes

EmbodiedOcc: Embodied 3D Occupancy Prediction for Vision-based Online Scene Understanding

UniDrive: Towards Universal Driving Perception Across Camera Configurations

Path Choice Matters for Clear Attributions in Path Methods

PlaneRAS: Learning Planar Primitives for 3D Plane Recovery

D3QE: Learning Discrete Distribution Discrepancy-aware Quantization Error for Autoregressive-Generated Image Detection

Authentic 4D Driving Simulation with a Video Generation Model

SpectralAR: Spectral Autoregressive Visual Generation

LowRankOcc: Tensor Decomposition and Low-Rank Recovery for Vision-based 3D Semantic Occupancy Prediction

SelfOcc: Self-Supervised Vision-Based 3D Occupancy Prediction

Segment Any Motion in Videos

Deep Adversarial Metric Learning

Hardness-Aware Deep Metric Learning

Deep Metric Learning via Adaptive Learnable Assessment

Deep Compositional Metric Learning

Dimension Embeddings for Monocular 3D Object Detection

Attributable Visual Similarity Learning

Tri-Perspective View for Vision-Based 3D Semantic Occupancy Prediction

Deep Factorized Metric Learning

Deep Relational Metric Learning

OPERA: Omni-Supervised Representation Learning with Hierarchical Supervisions

Token-Label Alignment for Vision Transformers

SurroundOcc: Multi-camera 3D Occupancy Prediction for Autonomous Driving

Structural Deep Metric Learning for Room Layout Estimation

GaussianWorld: Gaussian World Model for Streaming 3D Occupancy Prediction

Dynamic Metric Learning with Cross-Level Concept Distillation

GaussianFormer-2: Probabilistic Gaussian Superposition for Efficient 3D Occupancy Prediction

Learning Counterfactually Decoupled Attention for Open-World Model Attribution