Real-World Multiobject, Multigrasp Detection

Chu, Fu-Jen; Xu, Ruinian; Vela, Patricio A.

doi:10.1109/lra.2018.2852777

Cited by 324 publications

(252 citation statements)

References 30 publications

Supporting

Mentioning

245

Contrasting

Unclassified

Order By: Relevance

“…Therefore, the CGD could be selected as a suitable dataset for its quality and adaptability. The CGD was extensively used in [10,[18][19][20]31,33,35,50].…”

Section: Pre-compiled Datasetsmentioning

confidence: 99%

“…Zhou et al [34] used the ResNet-50 and ResNet-101 networks as feature extractors in their grasp detection work and achieved accuracies over 98% for both versions. Chu et al [50] used the same ResNet-50 architecture [70] with their grasp detection work. In contrast to the previous approach [10], they used grasp labels from the grasp data [60] to propose regions of interest in the images to ultimately propose multiple grasps at once.…”

Section: Architecturementioning

confidence: 99%

“…While arguing that the orientation predictions of a structured grasp representation lay in a non-Euclidean space, where standard regression loss (L 2 ) had not performed well, Chu et al [50] proposed to classify the orientation among 19 different classes in the range of [0 • , 360 • ]. They used a direct regression method to predict the bounding box of the grasp.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Review of Deep Learning Methods in Robotic Grasp Detection

Caldera

Rassau

Chai

2018

MTI

164

View full text Add to dashboard Cite

For robots to attain more general-purpose utility, grasping is a necessary skill to master. Such general-purpose robots may use their perception abilities to visually identify grasps for a given object. A grasp describes how a robotic end-effector can be arranged to securely grab an object and successfully lift it without slippage. Traditionally, grasp detection requires expert human knowledge to analytically form the task-specific algorithm, but this is an arduous and time-consuming approach. During the last five years, deep learning methods have enabled significant advancements in robotic vision, natural language processing, and automated driving applications. The successful results of these methods have driven robotics researchers to explore the use of deep learning methods in task-generalised robotic applications. This paper reviews the current state-of-the-art in regards to the application of deep learning methods to generalised robotic grasping and discusses how each element of the deep learning approach has improved the overall performance of robotic grasp detection. Several of the most promising approaches are evaluated and the most suitable for real-time grasp detection is identified as the one-shot detection method. The availability of suitable volumes of appropriate training data is identified as a major obstacle for effective utilisation of the deep learning approaches, and the use of transfer learning techniques is proposed as a potential mechanism to address this. Finally, current trends in the field and future potential research directions are discussed.

show abstract

“…Therefore, the CGD could be selected as a suitable dataset for its quality and adaptability. The CGD was extensively used in [10,[18][19][20]31,33,35,50].…”

Section: Pre-compiled Datasetsmentioning

confidence: 99%

Section: Architecturementioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Review of Deep Learning Methods in Robotic Grasp Detection

Caldera

Rassau

Chai

2018

MTI

164

View full text Add to dashboard Cite

show abstract

“…This approach is similar to MultiGrasp (as the family of YOLO object detectors is a direct descendant of Multi-Grasp [17]) with the key difference of predicting offsets to predefined anchor boxes for each grid cell, instead of directly predicting the best grasp at each cell. Chu et al [7] extends MultiGrasp to multiple object grasp detection by using region-of-interest pooling layers [23]. 4) Discrete Approaches: Johns et al [24] proposed to use a discretization of the space with a granularity of 1 cm and 30 • .…”

Section: A Proposal + Classification Approachesmentioning

confidence: 99%

GQ-STN: Optimizing One-Shot Grasp Detection based on Robustness Classifier

Gariépy¹,

Ruel²,

Chaib-draa³

et al. 2019

2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

Grasping is a fundamental robotic task needed for the deployment of household robots or furthering warehouse automation. However, few approaches are able to perform grasp detection in real time (frame rate). To this effect, we present Grasp Quality Spatial Transformer Network (GQ-STN), a oneshot grasp detection network. Being based on the Spatial Transformer Network (STN), it produces not only a grasp configuration, but also directly outputs a depth image centered at this configuration. By connecting our architecture to an externally-trained grasp robustness evaluation network, we can train efficiently to satisfy a robustness metric via the backpropagation of the gradient emanating from the evaluation network. This removes the difficulty of training detection networks on sparsely annotated databases, a common issue in grasping. We further propose to use this robustness classifier to compare approaches, being more reliable than the traditional rectangle metric. Our GQ-STN is able to detect robust grasps on the depth images of the Dex-Net 2.0 dataset with 92.4 % accuracy in a single pass of the network. We finally demonstrate in a physical benchmark that our method can propose robust grasps more often than previous sampling-based methods, while being more than 60 times faster.

show abstract

“…To evaluate the grasp detection performance of our GNet, we conduct real robotic grasp experiments on a UR5 robotic arm, by comparison with three typial grasp approaches, Hybrid Grasp [32], Multi-Modal Grasp [33] and Multi-Object Grasp [34].…”

Section: ) Gnet Based Physical Graspingmentioning

confidence: 99%

Learning Actions from Human Demonstration Video for Robotic Manipulation

Yang

Zhang

et al. 2019

2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

Learning actions from human demonstration is an emerging trend for designing intelligent robotic systems, which can be referred as video to command. The performance of such approach highly relies on the quality of video captioning. However, the general video captioning methods focus more on the understanding of the full frame, lacking of consideration on the specific object of interests in robotic manipulations. We propose a novel deep model to learn actions from human demonstration video for robotic manipulation. It consists of two deep networks, grasp detection network (GNet) and video captioning network (CNet). GNet performs two functions: providing grasp solutions and extracting the local features for the object of interests in robotic manipulation. CNet outputs the captioning results by fusing the features of both full frames and local objects. Experimental results on UR5 robotic arm show that our method could produce more accurate command from video demonstration than state-of-the-art work, thereby leading to more robust grasping performance.

show abstract

Real-World Multiobject, Multigrasp Detection

Cited by 324 publications

References 30 publications

Review of Deep Learning Methods in Robotic Grasp Detection

Review of Deep Learning Methods in Robotic Grasp Detection

GQ-STN: Optimizing One-Shot Grasp Detection based on Robustness Classifier

Learning Actions from Human Demonstration Video for Robotic Manipulation

Contact Info

Product

Resources

About