285 datasets found

D3Fields — dynamic 3D descriptor fields for zero-shot generalizable robotic manipulation tasks

The dataset used in the paper is RLBench, a standard benchmark for vision-based robotics which has been shown to serve as a proxy for real-robot experiments.

The Continual World benchmark consists of ten realistic robotic manipulation tasks.

Minigrid environment, a grid-based problem in reinforcement learning.

Synthetic datasets of transparent objects for training NeRF models.

REBEL is a dataset for reward regularization based robotic reinforcement learning from human feedback.

The dataset used in the paper is a collection of RGB images of a tabletop scene, along with their corresponding camera poses and 3D geometry.

The dataset used in this paper is a collection of 30 trajectories generated per gridworld map by randomly selecting for each trajectory the initial locations of both the actor...

The dataset used for training the Deep Kernel and Image Quality Estimators for optimizing the 6-dimensional robotic ultrasound controller.

A benchmark for evaluating task compositionality and long-horizon tasks through object manipulation, with language serving as the mechanism for goal specification.

The Cornell Grasp Detection Data Set consists of 1035 images of 280 different objects, with several images taken of each object by different orientations or poses.

The dataset used in the paper is the Mujoco benchmarking continuous control tasks.

The T-Less dataset is used for testing the class-adaptive object detector. It contains 15 objects with a large number of occurrences.

The HomeBrew dataset is used for testing the class-adaptive object detector. It contains 30 objects with a large number of occurrences.

The YCB-V dataset is used for testing the class-adaptive object detector. It contains 33 objects with a large number of occurrences.

Localization is one of the most important problems in various fields such as robotics and wireless communications. For instance, Unmanned Aerial Vehicles (UAVs) require the...

The dataset used in the paper is a set of demonstrations for a robot insertion task, which is a continuous control problem. The demonstrations are collected by a robot...

The RoboTHOR dataset is a simulation environment for embodied AI research, containing 15 validation environments with 12 different goal object categories.

A multi-directional high-resolution tactile sensor for robotic manipulation

The dataset used in the paper is a custom-made 3 degrees of freedom (DOF) hopping leg system, mounted on a vertical rail with 1 passive DOF and 2 active DOFs.