Real-time pixel-wise grasp affordance prediction based on multi-scale context information fusion

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Yongxiang Wu ◽  
Yili Fu ◽  
Shuguo Wang

Purpose This paper aims to use fully convolutional network (FCN) to predict pixel-wise antipodal grasp affordances for unknown objects and improve the grasp detection performance through multi-scale feature fusion. Design/methodology/approach A modified FCN network is used as the backbone to extract pixel-wise features from the input image, which are further fused with multi-scale context information gathered by a three-level pyramid pooling module to make more robust predictions. Based on the proposed unify feature embedding framework, two head networks are designed to implement different grasp rotation prediction strategies (regression and classification), and their performances are evaluated and compared with a defined point metric. The regression network is further extended to predict the grasp rectangles for comparisons with previous methods and real-world robotic grasping of unknown objects. Findings The ablation study of the pyramid pooling module shows that the multi-scale information fusion significantly improves the model performance. The regression approach outperforms the classification approach based on same feature embedding framework on two data sets. The regression network achieves a state-of-the-art accuracy (up to 98.9%) and speed (4 ms per image) and high success rate (97% for household objects, 94.4% for adversarial objects and 95.3% for objects in clutter) in the unknown object grasping experiment. Originality/value A novel pixel-wise grasp affordance prediction network based on multi-scale feature fusion is proposed to improve the grasp detection performance. Two prediction approaches are formulated and compared based on the proposed framework. The proposed method achieves excellent performances on three benchmark data sets and real-world robotic grasping experiment.

2021 ◽  
Vol 2078 (1) ◽  
pp. 012008
Author(s):  
Hui Liu ◽  
Keyang Cheng

Abstract Aiming at the problem of false detection and missed detection of small targets and occluded targets in the process of pedestrian detection, a pedestrian detection algorithm based on improved multi-scale feature fusion is proposed. First, for the YOLOv4 multi-scale feature fusion module PANet, which does not consider the interaction relationship between scales, PANet is improved to reduce the semantic gap between scales, and the attention mechanism is introduced to learn the importance of different layers to strengthen feature fusion; then, dilated convolution is introduced. Dilated convolution reduces the problem of information loss during the downsampling process; finally, the K-means clustering algorithm is used to redesign the anchor box and modify the loss function to detect a single category. The experimental results show that the improved pedestrian detection algorithm in the INRIA and WiderPerson data sets under different congestion conditions, the AP reaches 96.83% and 59.67%, respectively. Compared with the pedestrian detection results of the YOLOv4 model, the algorithm improves by 2.41% and 1.03%, respectively. The problem of false detection and missed detection of small targets and occlusion has been significantly improved.


Energies ◽  
2021 ◽  
Vol 14 (5) ◽  
pp. 1426
Author(s):  
Chuanyang Liu ◽  
Yiquan Wu ◽  
Jingjing Liu ◽  
Jiaming Han

Insulator detection is an essential task for the safety and reliable operation of intelligent grids. Owing to insulator images including various background interferences, most traditional image-processing methods cannot achieve good performance. Some You Only Look Once (YOLO) networks are employed to meet the requirements of actual applications for insulator detection. To achieve a good trade-off among accuracy, running time, and memory storage, this work proposes the modified YOLO-tiny for insulator (MTI-YOLO) network for insulator detection in complex aerial images. First of all, composite insulator images are collected in common scenes and the “CCIN_detection” (Chinese Composite INsulator) dataset is constructed. Secondly, to improve the detection accuracy of different sizes of insulator, multi-scale feature detection headers, a structure of multi-scale feature fusion, and the spatial pyramid pooling (SPP) model are adopted to the MTI-YOLO network. Finally, the proposed MTI-YOLO network and the compared networks are trained and tested on the “CCIN_detection” dataset. The average precision (AP) of our proposed network is 17% and 9% higher than YOLO-tiny and YOLO-v2. Compared with YOLO-tiny and YOLO-v2, the running time of the proposed network is slightly higher. Furthermore, the memory usage of the proposed network is 25.6% and 38.9% lower than YOLO-v2 and YOLO-v3, respectively. Experimental results and analysis validate that the proposed network achieves good performance in both complex backgrounds and bright illumination conditions.


2016 ◽  
Vol 12 (2) ◽  
pp. 126-149 ◽  
Author(s):  
Masoud Mansoury ◽  
Mehdi Shajari

Purpose This paper aims to improve the recommendations performance for cold-start users and controversial items. Collaborative filtering (CF) generates recommendations on the basis of similarity between users. It uses the opinions of similar users to generate the recommendation for an active user. As a similarity model or a neighbor selection function is the key element for effectiveness of CF, many variations of CF are proposed. However, these methods are not very effective, especially for users who provide few ratings (i.e. cold-start users). Design/methodology/approach A new user similarity model is proposed that focuses on improving recommendations performance for cold-start users and controversial items. To show the validity of the authors’ similarity model, they conducted some experiments and showed the effectiveness of this model in calculating similarity values between users even when only few ratings are available. In addition, the authors applied their user similarity model to a recommender system and analyzed its results. Findings Experiments on two real-world data sets are implemented and compared with some other CF techniques. The results show that the authors’ approach outperforms previous CF techniques in coverage metric while preserves accuracy for cold-start users and controversial items. Originality/value In the proposed approach, the conditions in which CF is unable to generate accurate recommendations are addressed. These conditions affect CF performance adversely, especially in the cold-start users’ condition. The authors show that their similarity model overcomes CF weaknesses effectively and improve its performance even in the cold users’ condition.


Sign in / Sign up

Export Citation Format

Share Document