Cap3D

此存储库托管用于使用预训练模型进行可扩展3D字幕的数据，包括Objaverse和ABO中3D对象的描述性字幕 (示例字幕)。它还包括Objaverse对象的点云和渲染图像，以及它们的shap-e潜在代码。此处包含的所有3D对象都具有商业友好许可证 (cc-by 4.0、cc-bby SA 4.0、CC0 1.0)。

This repository hosts data for scalable 3D captioning using pre-trained models, including descriptive captions (sample captions) for 3D objects sourced from both Objaverse and ABO. It also includes point clouds and rendered images of Objaverse objects, as well as their SHAP-E latent codes. All 3D objects contained herein are covered by commercially-friendly licenses: CC BY 4.0, CC BY-SA 4.0, and CC0 1.0.

搜集汇总

背景与挑战

背景概述

Cap3D是一个用于3D对象字幕处理的商业友好数据集，包含描述性字幕、点云、渲染图像及shap-e潜在代码，适用于预训练模型的研究与应用。