five

Downsampled Disentanglement Datasets - Falcor3D and Isaac3D

收藏
NIAID Data Ecosystem2026-03-11 收录
下载链接:
https://zenodo.org/record/3669343
下载链接
链接失效反馈
官方服务:
资源简介:
New disentanglement datasets This data repository contains the Falcor3D and Isaac3D datasets for disentanglement learning, where the image resolution is 128x128. Falcor3D The Falcor3D dataset consists of 233,280 images based on the 3D scene of a living room. The meta code corresponds to all possible combinations of 7 factors of variation: lighting_intensity (5) lighting_x-dir (6) lighting_y-dir (6) lighting_z-dir (6) camera_x-pos (6) camera_y-pos (6) camera_z-pos (6) Note that the number m behind each factor represents that the factor has m possible values, uniformly sampled in the normalized range of variations [0, 1]. Each image has as filename padded_index.png where index = lighting_intensity * 46656 + lighting_x-dir * 7776 + lighting_y-dir * 1296 + lighting_z-dir * 216 + camera_x-pos * 36 + camera_y-pos * 6 + camera_z-pos padded_index = index padded with zeros such that it has 6 digits. Isaac3D The Isaac3D dataset consists of 737,280 images, based on the 3D scene of a kitchen. The meta code corresponds to all possible combinations of 9 factors of variation: object_shape (3) object_scale (4) camera_height (4) robot_x-movement (8) robot_y-movement (5) lighting_intensity (4) lighting_y-dir (6) object_color (4) wall_color (4) Similarly, the number m behind each factor represents that the factor has m possible values, uniformly sampled in the normalized range of variations [0, 1]. Each image has as filename padded_index.png where index = object_shape * 245760 + object_scale * 30720 + camera_height * 6144 + robot_x-movement * 1536 + robot_y-movement * 384 + lighting_intensity * 96 + lighting_y-dir * 16 + object_color * 4 + wall color padded_index = index padded with zeros such that it has 6 digits.

新型解耦学习数据集(New disentanglement datasets) 本数据集仓库包含用于解耦学习(disentanglement learning)的Falcor3D与Isaac3D数据集,图像分辨率均为128×128。 Falcor3D数据集 该数据集基于客厅3D场景构建,共包含233280张图像。其元代码对应7个变异因子的全部可能组合: - 光照强度(lighting_intensity):共5个可选值 - 光照X方向(lighting_x-dir):共6个可选值 - 光照Y方向(lighting_y-dir):共6个可选值 - 光照Z方向(lighting_z-dir):共6个可选值 - 相机X位置(camera_x-pos):共6个可选值 - 相机Y位置(camera_y-pos):共6个可选值 - 相机Z位置(camera_z-pos):共6个可选值 注:每个因子后的数字m代表该因子共有m个可选值,均在归一化变异范围[0, 1]内均匀采样。 每张图像的文件名为`padded_index.png`,其中: index = lighting_intensity × 46656 + lighting_x-dir × 7776 + lighting_y-dir × 1296 + lighting_z-dir × 216 + camera_x-pos × 36 + camera_y-pos × 6 + camera_z-pos padded_index为将index补零至6位的结果。 Isaac3D数据集 该数据集基于厨房3D场景构建,共包含737280张图像。其元代码对应9个变异因子的全部可能组合: - 物体形状(object_shape):共3个可选值 - 物体缩放(object_scale):共4个可选值 - 相机高度(camera_height):共4个可选值 - 机器人X方向移动(robot_x-movement):共8个可选值 - 机器人Y方向移动(robot_y-movement):共5个可选值 - 光照强度(lighting_intensity):共4个可选值 - 光照Y方向(lighting_y-dir):共6个可选值 - 物体颜色(object_color):共4个可选值 - 墙面颜色(wall_color):共4个可选值 同理,每个因子后的数字m代表该因子共有m个可选值,均在归一化变异范围[0, 1]内均匀采样。 每张图像的文件名为`padded_index.png`,其中: index = object_shape × 245760 + object_scale × 30720 + camera_height × 6144 + robot_x-movement × 1536 + robot_y-movement × 384 + lighting_intensity × 96 + lighting_y-dir × 16 + object_color × 4 + wall_color padded_index为将index补零至6位的结果。
创建时间:
2020-02-17
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