Scholar
Yongming Rao
Google Scholar ID: 3qO6gK4AAAAJ
Tencent Hunyuan
Citations & Impact
Citations
11,583
H-index
41
i10-index
61
Publications
20
Co-authors
25
list available
Publications
Resume (English only)
Academic Achievements
- Published 'Unleashing Text-to-Image Diffusion Models for Visual Perception' at ICCV 2023, proposing the VPD framework, ranked 1st on NYUv2 Depth Estimation.
- Published 'HorNet: Efficient High-Order Spatial Interactions with Recursive Gated Convolutions' at NeurIPS 2022, introducing the HorNet vision backbone.
- Published 'P2P: Tuning Pre-trained Image Models for Point Cloud Analysis with Point-to-Pixel Prompting' at NeurIPS 2022, presenting the P2P framework for point cloud analysis.
- Published 'DenseCLIP: Language-Guided Dense Prediction with Context-Aware Prompting' at CVPR 2022, proposing the DenseCLIP framework for dense prediction.
- Published 'Point-BERT: Pre-Training 3D Point Cloud Transformers with Masked Point Modeling' at CVPR 2022, introducing unsupervised pre-training for 3D point cloud Transformers.
- Published 'Global Filter Networks for Image Classification' at NeurIPS 2021, proposing a frequency-domain transformer-style architecture.
- Published 'DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification' at NeurIPS 2021, presenting a dynamic token sparsification method.
- Published 'PoinTr: Diverse Point Cloud Completion with Geometry-Aware Transformers' at ICCV 2021 (Oral Presentation), reformulating point cloud completion as set-to-set translation.
- Published 'RandomRooms: Unsupervised Pre-training from Synthetic Shapes and Randomized Layouts for 3D Object Detection' at ICCV 2021.
