Point-MGE(ICME 2025 Oral):Point-MGE 深度融合 3D 点云的表示学习和生成学习。该框架首先使用向量量化变分自编码器(VQVAE)重建 3D 形状的神经场表示,从而学习点云补丁的离散语义特征。然后,设计了一个滑动掩蔽比例来平滑从表示学习到生成学习的过渡。
论文发表
Teqiang Zou*, Hongliang Zeng*, et al. "Asynchronous Fast-Slow Vision-Language-Action Policies for Whole-Body Robotic Manipulation." arXiv preprint, 2025. (共同一作)
Hongliang Zeng, Ping Zhang, et al. "Masked Generative Extractor for Synergistic Representation and 3D Generation of Point Clouds." ICME, 2025. Oral
Hongliang Zeng, Ping Zhang, et al. "GMAP: Generalized Manipulation of Articulated Objects in Robotic Using Pre-trained Model." AAAI, 2025. Oral
Hongliang Zeng, Ping Zhang, et al. "Point-UMAE: Unet-like Masked Autoencoders for Point Cloud Self-supervised Learning." ICASSP, 2025. Oral
Hongliang Zeng, Ping Zhang, et al. "Active Visual Learning for Robots with Dueling Deep Q-Networks and Transformer Encoders." ICASSP, 2025.
Hongliang Zeng, Ping Zhang, et al. "MARS: Multimodal Active Robotic Sensing for Articulated Characterization." IJCAI, 2024. Oral
Hongliang Zeng, Ping Zhang, et al. "AHEGC: Adaptive Hindsight Experience Replay With Goal-Amended Curiosity Module for Robot Control." TNNLS, 2024. (JCR Q1, IF: 10.4)
Jiahua Wang, Ping Zhang, Hongliang Zeng, et al. "FAEMTrack: Feature-Augmented Embedding and Cross-Drone Fusion for Single Object Tracking." IEEE Internet of Things Journal, 2025.
Tingyu Ye, Ping Zhang, Hongliang Zeng, et al. "Multi-objective Neural Architecture Search Combining Binary Artificial Bee Colony Algorithm for Dynamic Hand Gesture Recognition." Expert Systems with Applications, 2025.
Tingyu Ye, Ping Zhang, Hui Wang, Hongliang Zeng, et al. "Reinforcement Learning-Driven Dual Neighborhood Structure Artificial Bee Colony Algorithm for Continuous Optimization Problem." Applied Soft Computing, 2024.