Generative Robotic Grasping Using Depthwise Separable Convolution

Teng, Yadong; Gao, Peng

doi:10.1016/j.compeleceng.2021.107318

Cited by 19 publications

(8 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the inverted residual structure block, the ordinary convolution is rep depth-wise separable convolution. Compared with the ordinary convolution wise separable convolution first performs the channel-by-channel convolutio extraction and then performs the point-by-point volume of the expansion c product [28] is shown in Figure 5; under the premise of ensuring the numb map channels, the number of parameters and amount of computation are grea In the inverted residual structure block, the ordinary convolution is replaced by the depth-wise separable convolution. Compared with the ordinary convolution, the depthwise separable convolution first performs the channel-by-channel convolution of feature extraction and then performs the point-by-point volume of the expansion channel.…”

Section: Experiments Methodsmentioning

confidence: 99%

“…Compared with the ordinary convolution, the depthwise separable convolution first performs the channel-by-channel convolution of feature extraction and then performs the point-by-point volume of the expansion channel. The product [28] is shown in Figure 5; under the premise of ensuring the number of feature map channels, the number of parameters and amount of computation are greatly reduced. Taking the input image as size 𝐷 × 𝐷 × 𝑀, and the convolution kern 𝐷 × 𝐷 × 𝑀, for example, the number of output channels is N, and the calcu of ordinary convolution is:…”

Section: Experiments Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Lightweight Yolov4 Target Detection Algorithm Fused with ECA Mechanism

Wang

Zhou

2022

Processes

View full text Add to dashboard Cite

For the task of garbage classification, to overcome the main disadvantages of the Yolov4 target detection algorithm, such as the large network model and lower detection accuracy for small objects, a lightweight Yolov4 target detection network based on the EfficientNet-B0 fusion ECA mechanism is presented. The lightweight EfficientNet was used to replace the original backbone network, which reduces the parameters of the network model and improves the detection accuracy. Moreover, a deep detachable convolution block replaced the common convolution block in the original network, which further reduced the number of parameters in the model. In the feature pyramid model PANet, a lightweight ECA attention mechanism was introduced to realize the weight analysis of the importance of different channel feature maps through cross-channel interaction, allowing the network to extract more obvious features with which to distinguish categories. Finally, a soft-NMS algorithm was introduced in the post-processing stage of the detection frame to reduce the missed target detection rate in dense areas, which can improve the detection accuracy of the network and detection efficiency. As shown in the results, the size of the model was only 48 MB, and the mAP was 91.09%. Compared with the original Yolov4 network, the mAP was increased by 5.77% based on the 80% reduction in the model size. The recognition of small targets was also improved, which proved the effectiveness and robustness of the improved algorithm.

show abstract

Section: Experiments Methodsmentioning

confidence: 99%

Section: Experiments Methodsmentioning

confidence: 99%

Lightweight Yolov4 Target Detection Algorithm Fused with ECA Mechanism

Wang

Zhou

2022

Processes

View full text Add to dashboard Cite

show abstract

“…This method achieved accuracies of 97.8% on the Cornell dataset and 95.6% on the Jacquard dataset. Teng et al [7] utilized depth-wise separable convolutions to make a grasping network more lightweight. Fu et al [22] also utilized depth-wise separable convolution to create a lightweight network.…”

Section: Deep Learning For Robot Graspingmentioning

confidence: 99%

“…Instead, we separately output sin(2θ) and cos(2θ) to eliminate discontinuities near ± π a unique mapping between θ and the interval − π 2 , π 2 . Finally, the grasping angle is generated using Equation (7).…”

Section: The Proposed Fagd-net For Grasping Detectionmentioning

confidence: 99%

See 1 more Smart Citation

FAGD-Net: Feature-Augmented Grasp Detection Network Based on Efficient Multi-Scale Attention and Fusion Mechanisms

Zhong,

Liu,

Gong

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

Grasping robots always confront challenges such as uncertainties in object size, orientation, and type, necessitating effective feature augmentation to improve grasping detection performance. However, many prior studies inadequately emphasize grasp-related features, resulting in suboptimal grasping performance. To address this limitation, this paper proposes a new grasping approach termed the Feature-Augmented Grasp Detection Network (FAGD-Net). The proposed network incorporates two modules designed to enhance spatial information features and multi-scale features. Firstly, we introduce the Residual Efficient Multi-Scale Attention (Res-EMA) module, which effectively adjusts the importance of feature channels while preserving precise spatial information within those channels. Additionally, we present a Feature Fusion Pyramidal Module (FFPM) that serves as an intermediary between the encoder and decoder, effectively addressing potential oversights or losses of grasp-related features as the encoder network deepens. As a result, FAGD-Net achieved advanced levels of grasping accuracy, with 98.9% and 96.5% on the Cornell and Jacquard datasets, respectively. The grasp detection model was deployed on a physical robot for real-world grasping experiments, where we conducted a series of trials in diverse scenarios. In these experiments, we randomly selected various unknown household items and adversarial objects. Remarkably, we achieved high success rates, with a 95.0% success rate for single-object household items, 93.3% for multi-object scenarios, and 91.0% for cluttered scenes.

show abstract

Grasping Point Detection Using Monocular Camera Image Processing and Knowledge of Center of Gravity

Štursa

Doležel

Honc

2022

Artificial Intelligence Trends in Systems

View full text Add to dashboard Cite

The ability to grasp objects is one of the basic functions of modern industrial robots. In this article, the focus is placed on a system for processing the image provided by a robot visual perception system leading to the detection of objects grasping points. The proposed processing system is based on a multi-step method using convolutional neural networks (CNN). The first step is to use the first CNN to transform the input image into a schematic image with labeled objects centers of gravity, which then serves as a supporting input to the second CNN. In this second CNN, original input and supporting input images are used to obtain a schematic image containing the grasping points of the objects. This solution is further compared with a network providing grasping points directly from the input image. As a result, the proposed method provided a 0.7% improvement in the average intersection over union for all of the models.

show abstract

Generative Robotic Grasping Using Depthwise Separable Convolution

Cited by 19 publications

References 7 publications

Lightweight Yolov4 Target Detection Algorithm Fused with ECA Mechanism

Lightweight Yolov4 Target Detection Algorithm Fused with ECA Mechanism

FAGD-Net: Feature-Augmented Grasp Detection Network Based on Efficient Multi-Scale Attention and Fusion Mechanisms

Grasping Point Detection Using Monocular Camera Image Processing and Knowledge of Center of Gravity

Contact Info

Product

Resources

About