Full name | A Unified Query-based Paradigm for Point Cloud Understanding |
Description | EQ-Net is a variant of SparseConv U-Net under the Embedding-Querying (EQ) paradigm. We develop EQ-Net with EQ-Paradigm based on the sparse U-Net by keeping the sparseconv encoder as our embedding network and adopting the Q-Net to extract the features for each point. |
Publication title | A Unified Query-based Paradigm for Point Cloud Understanding |
Publication authors | Zetong Yang*, Li Jiang*, Yanan Sun, Bernt Schiele, Jiaya JIa |
Publication venue | CVPR 2022 |
Publication URL | https://arxiv.org/pdf/2203.01252.pdf |
Input Data Types | Uses Color Uses 3D |
Programming language(s) | Python with CUDA |
Hardware | Titan Xp |
Source code or download URL | https://github.com/dvlab-research/DeepVision3D |
Submission creation date | 16 Nov, 2021 |
Last edited | 12 May, 2022 |
Last uploaded | 16 Nov, 2021 |