| Full name | PointGroup + Transformer (w/o end-to-end fine-tuning) |
| Description | This is the pretrained version of D3Net - PointGroup + Transformer (w/o end-to-end fine-tuning) |
| Publication title | D3Net: A Unified Speaker-Listener Architecture for 3D Dense Captioning and Visual Grounding |
| Publication authors | Dave Zhenyu Chen, Qirui Wu, Matthias Niessner, Angel X. Chang |
| Publication venue | 17th European Conference on Computer Vision (ECCV), 2022 |
| Publication URL | https://arxiv.org/abs/2112.01551 |
| Input Data Types | Uses XYZ coordinates,Uses Multiview Image Features,Uses Normal Vectors |
| Programming language(s) | Python |
| Hardware | RTX 3090Ti |
| Website | https://daveredrum.github.io/D3Net/ |
| Source code or download URL | https://github.com/daveredrum/D3Net |
| Submission creation date | 27 Oct, 2021 |
| Last edited | 23 Jul, 2022 |