Metric learning based object recognition and retrieval

Yang, Jianyu; Xu, Haoran

doi:10.1016/j.neucom.2016.01.032

Cited by 15 publications

(5 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Xiong et al [ 27 ] proposed the extraction of features from different viewpoints using the anchors for 3D pose estimation. Although the discriminative methods can directly regress the 3D human pose without the time-costly process of fitting the complicated human model to the depth image, the performance of these methods on self-occlusion [ 37 , 38 ] human poses is poor.…”

Section: Related Workmentioning

confidence: 99%

WPL-Based Constraint for 3D Human Pose Estimation from a Single Depth Image

Xing

Yang

2022

Sensors

Self Cite

View full text Add to dashboard Cite

Three-dimensional human pose estimation from depth maps is a fast-growing research area in computer vision. The distal joints of the human body are more flexible than the proximal joints, making it more difficult to estimate the distal joints. However, most existing methods ignore the difference between the distal joints and proximal joints. Moreover, the distal joint can be constrained by the proximal joint on the same kinematic chain. In our work, we model the human skeleton as the tree structure called the human-tree. Then, motivated by the WPL (weighted path length) in the data structure, we propose a WPL-based loss function to constrain the distal joints with the proximal joints in a global-to-local manner. Extensive experiments on benchmarks demonstrate that our method can effectively improve the performance of the distal joints.

show abstract

Section: Related Workmentioning

confidence: 99%

WPL-Based Constraint for 3D Human Pose Estimation from a Single Depth Image

Xing

Yang

2022

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…Most recent studies with RCA involve the inclusion of cannotlinks such as in [18] and [19] and kernelization in order to allow handling of nonlinear metrics, such as in [20]. RCA-like approaches have also seen success in several applications, notably in image pattern recognition, as in [21] and [22].…”

Section: Weakly Supervised Feature Space Transformationmentioning

confidence: 99%

Weakly Supervised Acoustic Defect Detection in Concrete Structures Using Clustering-Based Augmentation

Kasahara

Fujii

Yamashita

et al. 2021

IEEE/ASME Trans. Mechatron.

View full text Add to dashboard Cite

The automation of inspection methods for concrete structures is a pressing issue worldwide. Weakly supervised approaches, i.e., approaches based on supervision in other forms than traditional class labels, offer a unique mix of automation and human involvement that is highly effective for critical tasks such as inspection work. Generating weak supervision is less tedious than generating training data for supervised learning approaches. However, since it is less informative, high amounts of weak supervision are often needed. In practice, it is often the case that only scarce amounts of weak supervision are available. In this article, we propose a novel approach for weakly supervised acoustic defect detection in concrete structures that augment human-provided weak supervision. Experiments in both laboratory and field conditions showed that the proposed method allows for considerable performance gains for low amounts of weak supervision.

show abstract

“…RCA is a metric learning clustering method proposed initially in [16]. Several variants exists [19][20][21] and recently several successful applications have been reported [22,23].…”

Section: Algorithmmentioning

confidence: 99%

Acoustic Inspection of Concrete Structures Using Active Weak Supervision and Visual Information

Kasahara

Yamashita

Asama

2020

Sensors

View full text Add to dashboard Cite

Concrete structures are featured heavily in most modern societies. In recent years, the need to inspect those structures has been a growing concern and the automation of inspection methods is highly demanded. Acoustic methods such as the hammering test are one of the most popular non-destructive testing methods for this task. In this paper, an approach to defect detection in concrete structures with active weak supervision and visual information is proposed. Based on audio and position information, pairs of samples are actively queried to a user on their similarity. Those are used to transform the feature space into a favorable one, in a weakly supervised fashion, for clustering defect and non-defect samples, reinforced by position information. Experiments conducted in both laboratory conditions and in field conditions proved the effectiveness of the proposed method.

show abstract

Metric learning based object recognition and retrieval

Cited by 15 publications

References 42 publications

WPL-Based Constraint for 3D Human Pose Estimation from a Single Depth Image

WPL-Based Constraint for 3D Human Pose Estimation from a Single Depth Image

Weakly Supervised Acoustic Defect Detection in Concrete Structures Using Clustering-Based Augmentation

Acoustic Inspection of Concrete Structures Using Active Weak Supervision and Visual Information

Contact Info

Product

Resources

About