Efficient grasping from RGBD images: Learning using a new rectangle representation

By Yun Jiang, Stephen Moseson, Ashutosh Saxena. 2011

723 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

Learning human activities and object affordances from RGB-D videos (2012, background): “Some recent works (Gupta et al., 2009; Yao and Fei-Fei, 2010; Aksoy et al., 2010; Jiang et al., 2011a; Pirsiavash and Ramanan, 2012) show that modeling the interaction between human poses and objects in 2D videos results in a better performance on the tasks of object detection and activity…”
Supersizing self-supervision: Learning to grasp from 50K tries and 700 robot hours (2015, methodology): “This ensures that in the next iteration, the robot corrects for incorrect grasp modalities while reinforcing the correct ones.”
Deep learning for detecting robotic grasps (2013, methodology): “Sensing for Grasping: In a real-world robotic setting, a robot will not have full knowledge of the 3D model and pose of an object to be grasped, but rather only incomplete information from some set of sensors such as color or depth cameras, tactile sensors, etc.”
Real-time grasp detection using convolutional neural networks (2014, background): “Visual models of grasps generalize well to novel objects and only require a single view of the object, not a full physical model [14] [2].”
GraspNet-1Billion: A Large-Scale Benchmark for General Object Grasping (2020, background): “Moreover, most of the datasets adopt the rectangle based representation [16] of grasp pose, which constrains the space for placing the gripper.”
Deep-learning in Mobile Robotics - from Perception to Control Systems: A Survey on Why and Why not (2016, background): “Jiang et al. (2011) and Zhang et al. (2011) converted that to a detection problem.”
Deep learning a grasp function for grasping under gripper pose uncertainty (2016, methodology): “The Cornell grasping dataset [7] for example, consists of RGB and depth images with a parallel-jaw gripper pose defined in image coordinates.”
Jacquard: A Large Scale Dataset for Robotic Grasp Detection (2018, neutral): “dimensions described in [13] but Lenz et al.”
Learning robust, real-time, reactive robotic grasping (2019, methodology): “Both represent antipodal grasps in RGB and depth images as rectangles using pixel coordinates (Yun Jiang et al., 2011).”
Grasp Pose Detection in Point Clouds (2017, methodology): “For example, Jiang et al. models a graspable geometry as an oriented rectangle in an RGBD image (Jiang et al. 2011).”
Real-World Multiobject, Multigrasp Detection (2018, background): “The 5-dimensional grasp rectangle is the grasp representation employed [15].”
More Than a Feeling: Learning to Grasp and Regrasp Using Vision and Touch (2018, positive): “As an alternative, data-driven approaches have sought to predict grasp outcomes from human supervision [19, 29, 14, 22], simulation [20, 16, 25], or autonomous robotic data collection [26, 23], typically using visual or depth observations.”
Vision-based robotic grasping from object localization, object pose estimation to grasp estimation for parallel grippers: a review (2020, methodology): “Comparisons The methods of evaluating oriented rectangles are compared in Table 11 on the widely used Cornell Grasping dataset [Jiang et al., 2011].”
Aligning Cyber Space With Physical World: A Comprehensive Survey on Embodied AI (2024, neutral): “2) Simulated Data: Data collection in real-world settings is resource-intensive and time-consuming, making simulation-based data collection an attractive alternative.”
Hallucinated Humans as the Hidden Context for Labeling 3D Scenes (2013, background): “Reasoning in 3D allows an algorithm to capture stronger context, such as shape, stability and orientation of the objects [15, 13, 16].”