Wenjun Zeng

10

Papers

70

Total Citations

Papers (10)

Hierarchical Temporal Context Learning for Camera-based Semantic Scene Completion

Closed-Loop Unsupervised Representation Disentanglement with $\beta$-VAE Distillation and Diffusion Probabilistic Feedback

Rate-Distortion-Cognition Controllable Versatile Neural Image Compression

One at a Time: Progressive Multi-Step Volumetric Probability Learning for Reliable 3D Scene Perception

Disentangled World Models: Learning to Transfer Semantic Knowledge from Distracting Videos for Reinforcement Learning

Perceiving and Acting in First-Person: A Dataset and Benchmark for Egocentric Human-Object-Human Interactions

ULTHO: Ultra-Lightweight yet Efficient Hyperparameter Optimization in Deep Reinforcement Learning

ReGenNet: Towards Human Action-Reaction Synthesis

Inter-X: Towards Versatile Human-Human Interaction Analysis

UniScene: Unified Occupancy-centric Driving Scene Generation