five

cun-bjy/mpi3d_real

收藏
Hugging Face2023-09-05 更新2024-03-04 收录
下载链接:
https://hf-mirror.com/datasets/cun-bjy/mpi3d_real
下载链接
链接失效反馈
官方服务:
资源简介:
--- license: bsd task_categories: - feature-extraction language: - ar tags: - code pretty_name: mpi3d_real size_categories: - 100M<n<1B --- # mpi3d_real This repository is a _unofficial_ backup service for `MPI3D-Real` dataset, which is provided in this paper: [**On the Transfer of Inductive Bias from Simulation to the Real World: a New Disentanglement Dataset**](https://proceedings.neurips.cc/paper/2019/hash/d97d404b6119214e4a7018391195240a-Abstract.html). The dataset contains images of real objects with varying factors of variation, such as shape, color, texture and pose. The dataset is useful for studying the generalization and disentanglement abilities of representation learning models. The backup service allows users to download the dataset from a mirror site in case the original source is unavailable. For more detailed information on the dataset, please check the [original repository](https://github.com/rr-learning/disentanglement_dataset) ## Reference [1] Gondal, Muhammad Waleed, et al. "On the transfer of inductive bias from simulation to the real world: a new disentanglement dataset." Advances in Neural Information Processing Systems 32 (2019).
提供机构:
cun-bjy
原始信息汇总

mpi3d_real

数据集概述

  • 名称: mpi3d_real
  • 许可证: BSD
  • 任务类别: 特征提取
  • 语言: 阿拉伯语
  • 标签: code
  • 大小类别: 100M<n<1B
  • 别名: mpi3d_real

数据集描述

  • 来源: 非官方备份服务,用于MPI3D-Real数据集。
  • 内容: 包含真实物体的图像,具有形状、颜色、纹理和姿态等多种变化因素。
  • 用途: 适用于研究表示学习模型的泛化和解纠缠能力。
  • 备份服务: 允许用户从镜像站点下载数据集,以防原始来源不可用。

参考文献

  • 文献: Gondal, Muhammad Waleed, et al. "On the transfer of inductive bias from simulation to the real world: a new disentanglement dataset." Advances in Neural Information Processing Systems 32 (2019).
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作