hgupt3/sitr_dataset

SITR（Sensor-Invariant Tactile Representation）数据集包含了模拟和真实世界设置下用于训练和评估传感器不变性触觉表示的模型的数据。数据集分为三个主要部分：模拟触觉数据集、分类触觉数据集和位姿估计触觉数据集。模拟触觉数据集是通过Blender中的基于物理的渲染生成的，包含100种独特的模拟传感器配置。分类触觉数据集和位姿估计触觉数据集分别来自真实世界的7种传感器和4种传感器的数据。

The SITR (Sensor-Invariant Tactile Representation) dataset includes data for training and evaluating models for sensor-invariant tactile representations across simulated and real-world settings. The dataset is divided into three main parts: the Simulated Tactile Dataset, the Classification Tactile Dataset, and the Pose Estimation Tactile Dataset. The Simulated Tactile Dataset is generated using physics-based rendering in Blender and consists of 100 unique simulated sensor configurations. The Classification Tactile Dataset and Pose Estimation Tactile Dataset come from data collected using 7 real-world sensors and 4 sensors, respectively.