Yale-CMU-Berkeley dataset for robotic manipulation research

By Berk Çalli, Arjun Singh, James Bruce, Aaron Walsman, Kurt Konolige, Siddhartha S. Srinivasa, Pieter Abbeel, Aaron M. Dollar. 2017

453 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Control and Systems Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Control and Systems Engineering

Citations in the corpus

Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data (2019, background): “Some datasets [36, 5] are captured in controlled environment which might greatly differ from real-world scenes.”
SAPIEN: A SimulAted Part-Based Interactive ENvironment (2020, methodology): “We repeat this experiment with PointNet++[4] operating on 3D RGB-point cloud produced by the same images.”
Keypoints into the Future: Self-Supervised Correspondence in Model-Based Reinforcement Learning (2020, background): “• top down camera: This environment, depicted in Figure 3 (a), consists of the sugar-box object from the YCB dataset [3] • occlusions: This environment uses the same setup of task angled camera the only difference being that the object is now laying on its side, see Figure 3 (c).”
6-DOF Grasping for Target-driven Object Manipulation in Clutter (2019, methodology): “t object is ﬁnally grasped in (f). Fig. 9: Scenes used for testing. See accompanying video for grasp performance. varying geometry, physical and visual properties, including some from the YCB dataset [45]. We compared the performance of the method on 9 different scenes (see Fig9) with the ﬁxed order”
Segmentation-Driven 6D Object Pose Estimation (2018, background): “It comprises 21 objects taken from the YCB dataset [5, 4], which are of diverse sizes and with different degrees of texture.”
GraspNet-1Billion: A Large-Scale Benchmark for General Object Grasping (2020, methodology): “We select 32 objects that are suitable for grasping from the YCB dataset [6], 13 adversarial objects from DexNet 2.”
DexVIP: Learning Dexterous Grasping with Human Hand Pose Priors from Video (2022, methodology): “We acquire 3D object models from multiple public repositories: ContactDB [37], 3DNet [54], YCB [55], Free 3D [56], and 3D Warehouse [57].”
Rearrangement: A Challenge for Embodied AI (2020, methodology): “Evaluation could be based on standard object datasets, similar to the YCB dataset [12].”
Grasp'D: Differentiable Contact-rich Grasp Synthesis for Multi-fingered Hands (2022, methodology): “Since analytic synthesized datasets do not include many high-contact grasps, the authors also released the YCB Affordance dataset of 367 fine-grained grasps of the YCB object set [10], created by manually setting MANO hand joint angles in the GraspIt! simulator’s GUI.”
Language Grounding with 3D Objects (2021, unknown): “Prior work has gathered language annotations for the YCB Benchmark object set [38] to explore how language descriptions provide priors on object affordances [39].”
MaskFusion: Real-Time Recognition, Tracking and Reconstruction of Multiple Moving Objects (2018, methodology): “We conducted a quantitative evaluation of the quality of the 3D reconstruction achieved by MaskFusion using objects from the YCB Object and Model Set [4], a benchmark designed to facilitate progress in robotic manipulation applications.”