Real-time pose estimation of rigid objects in heavily cluttered environments

2015 ◽  
Vol 141 ◽  
pp. 38-51 ◽  
Author(s):  
Blaž Bratanič ◽  
Franjo Pernuš ◽  
Boštjan Likar ◽  
Dejan Tomaževič
Keyword(s):  
Real Time ◽  
Pose Estimation ◽  
Cluttered Environments ◽  
Rigid Objects

scholarly journals Faster than Real-Time Surface Pose Estimation with Application to Autonomous Robotic Grasping

Robotics ◽  
2022 ◽  
Vol 11 (1) ◽  
pp. 7
Author(s):  
Yannick Roberts ◽  
Amirhossein Jabalameli ◽  
Aman Behal
Keyword(s):  
Real Time ◽  
Pose Estimation ◽  
First Principles ◽  
Depth Map ◽  
Software Implementation ◽  
Grasp Planning ◽  
Cluttered Environments ◽  
Novel Approach ◽  
Object Rotation ◽  
Six Dof

Motivated by grasp planning applications within cluttered environments, this paper presents a novel approach to performing real-time surface segmentations of never-before-seen objects scattered across a given scene. This approach utilizes an input 2D depth map, where a first principles-based algorithm is utilized to exploit the fact that continuous surfaces are bounded by contours of high gradient. From these regions, the associated object surfaces can be isolated and further adapted for grasp planning. This paper also provides details for extracting the six-DOF pose for an isolated surface and presents the case of leveraging such a pose to execute planar grasping to achieve both force and torque closure. As a consequence of the highly parallel software implementation, the algorithm is shown to outperform prior approaches across all notable metrics and is also shown to be invariant to object rotation, scale, orientation relative to other objects, clutter, and varying degree of noise. This allows for a robust set of operations that could be applied to many areas of robotics research. The algorithm is faster than real time in the sense that it is nearly two times faster than the sensor rate of 30 fps.

scholarly journals Robust hand gesture recognition using multiple shape-oriented visual cues

2021 ◽  
Vol 2021 (1) ◽  
Author(s):  
Samy Bakheet ◽  
Ayoub Al-Hamadi
Keyword(s):  
Real Time ◽  
Gesture Recognition ◽  
Pose Estimation ◽  
Depth Map ◽  
Hand Gesture Recognition ◽  
Support Vector ◽  
Hand Gesture ◽  
Hand Pose Estimation ◽  
Time Operation ◽  
Hand Pose

AbstractRobust vision-based hand pose estimation is highly sought but still remains a challenging task, due to its inherent difficulty partially caused by self-occlusion among hand fingers. In this paper, an innovative framework for real-time static hand gesture recognition is introduced, based on an optimized shape representation build from multiple shape cues. The framework incorporates a specific module for hand pose estimation based on depth map data, where the hand silhouette is first extracted from the extremely detailed and accurate depth map captured by a time-of-flight (ToF) depth sensor. A hybrid multi-modal descriptor that integrates multiple affine-invariant boundary-based and region-based features is created from the hand silhouette to obtain a reliable and representative description of individual gestures. Finally, an ensemble of one-vs.-all support vector machines (SVMs) is independently trained on each of these learned feature representations to perform gesture classification. When evaluated on a publicly available dataset incorporating a relatively large and diverse collection of egocentric hand gestures, the approach yields encouraging results that agree very favorably with those reported in the literature, while maintaining real-time operation.

scholarly journals Real-Time Human Pose Estimation and Gesture Recognition from Depth Images Using Superpixels and SVM Classifier

Sensors ◽  
2015 ◽  
Vol 15 (6) ◽  
pp. 12410-12427 ◽  
Author(s):  
Hanguen Kim ◽  
Sangwon Lee ◽  
Dongsung Lee ◽  
Soonmin Choi ◽  
Jinsun Ju ◽  
...  
Keyword(s):  
Real Time ◽  
Gesture Recognition ◽  
Pose Estimation ◽  
Human Pose Estimation ◽  
Svm Classifier ◽  
Depth Images ◽  
Human Pose

Safe manipulation in unknown, crowded environments via sensor-based interleaving planner: interleaving software and sensitive skin hardware

Robotica ◽  
2016 ◽  
Vol 35 (5) ◽  
pp. 1176-1191
Author(s):  
Dugan Um ◽  
Dongseok Ryu
Keyword(s):  
Real Time ◽  
Configuration Space ◽  
Uncertain Environments ◽  
Sensitive Skin ◽  
Cluttered Environments ◽  
Time Planning ◽  
Near Future ◽  
Indoor Spaces ◽  
Crowded Environments ◽  
Complete Path

SUMMARYAs various robots are anticipated to coexist with humans in the near future, safe manipulation in unknown, cluttered environments becomes an important issue. Manipulation in an unknown environment, however, has been proven to be NP-Hard and the risk of unexpected human--robot collision hampers the dawning of the era of human--robot coexistence. We propose a non-contact-based sensitive skin as a means to provide safe manipulation hardware and interleaving planning between the workspace and the configuration space as software to solve manipulation problems in unknown, crowded environments. Novelty of the paper resides in demonstration of real time and yet complete path planning in an uncertain and crowded environment. To that end, we introduce the framework of the sensor-based interleaving planner (SBIP) whereby search completeness and safe manipulation are both guaranteed in cluttered environments. We study an interleaving mechanism between sensation in a workspace and execution in the corresponding configuration space for real-time planning in uncertain environments, thus the name interleaving planner implies.Applications of the proposed system include manipulators of a humanoid robot, surgical manipulators, and robotic manipulators working in hazardous and uncertain environments such as underwater, unexplored planets, and unstructured indoor spaces.

Sign in / Sign up

Export Citation Format

Share Document