Superpixel-Based Hand Gesture Recognition With Kinect Depth Camera

Wang, Chong; Liu, Zhong; Chan, S.C.

doi:10.1109/tmm.2014.2374357

Cited by 284 publications

(113 citation statements)

References 33 publications

Supporting

Mentioning

111

Contrasting

Unclassified

Order By: Relevance

“…While in a recommendation system, hand gesture can express "Like" or "Dislike" emotions of users. Thus, it is one of the most fundamental problems in computer vision and pattern recognition, and has a wide range of applications such as virtual reality systems [1], interactive gaming platforms [2], recognizing sign language [3,4,5], enabling very young children to interact with computers [6], controlling robot [7,8], practicing music conducting [9], television control [10,11], automotive interfaces [12,13], learning and teaching assistance [14,15], and hand gesture generation [16]. There has been significant progress in hand gesture recognition, however, some key problems e.g., fast and robust are still challenging.…”

Section: Introductionmentioning

confidence: 99%

“…Prior work usually puts emphasis on using whole data series, which always contain redundant information, resulting in degraded performance. For examples, Wang et al [1] present a superpixel-based hand gesture recognition system based on a novel superpixel earth mover's distance metric. Ren et al [2] focus on building a robust part-based hand gesture recognition system.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fast and robust dynamic hand gesture recognition via key frames extraction and feature fusion

Tang

Liu

Wei³

et al. 2019

Neurocomputing

109

View full text Add to dashboard Cite

Gesture recognition is a hot topic in computer vision and pattern recognition, which plays a vitally important role in natural human-computer interface. Although great progress has been made recently, fast and robust hand gesture recognition remains an open problem, since the existing methods have not well balanced the performance and the efficiency simultaneously. To bridge it, this work combines image entropy and density clustering to exploit the key frames from hand gesture video for further feature extraction, which can improve the efficiency of recognition. Moreover, a feature fusion strategy is also proposed to further improve feature representation, which elevates the performance of recognition. To validate our approach in a "wild" environment, we also introduce two new datasets called HandGesture and Action3D datasets. Experiments consistently demonstrate that our strategy achieves competitive results on Northwestern University, Cambridge, HandGesture and Action3D hand gesture datasets. Our code and datasets will release at https://github.com/Ha0Tang/HandGestureRecognition.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Fast and robust dynamic hand gesture recognition via key frames extraction and feature fusion

Tang

Liu

Wei³

et al. 2019

Neurocomputing

109

View full text Add to dashboard Cite

show abstract

“…Representing images as a non-overlapping set of superpixels is a standard practice as a pre-processing step for many computer vision applications, including depth estimation [12], localization [2], tracking [25], gesture recognition [23], human pose estimation [9], place recognition [15] and semantic segmentation [17]. By using superpixels instead of raw pixels, algorithms become more computationally efficient, with the added advantage that superpixels contain much more discriminative information than single pixels, e.g., color histograms and shape.…”

Section: Introductionmentioning

confidence: 99%

BASS: Boundary-Aware Superpixel Segmentation

Bandera

Yu²,

Simo-Serra

et al. 2016

2016 23rd International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

Abstract-We propose a new superpixel algorithm based on exploiting the boundary information of an image, as objects in images can generally be described by their boundaries. Our proposed approach initially estimates the boundaries and uses them to place superpixel seeds in the areas in which they are more dense. Afterwards, we minimize an energy function in order to expand the seeds into full superpixels. In addition to standard terms such as color consistency and compactness, we propose using the geodesic distance which concentrates small superpixels in regions of the image with more information, while letting larger superpixels cover more homogeneous regions. By both improving the initialization using the boundaries and coherency of the superpixels with geodesic distances, we are able to maintain the coherency of the image structure with fewer superpixels than other approaches. We show the resulting algorithm to yield smaller Variation of Information metrics in seven different datasets while maintaining Undersegmentation Error values similar to the state-of-the-art methods.

show abstract

“…Benzer bir çalışma daha sonra Wang ve diğ. tarafından önerilmiştir [5]. Süper piksel yeryüzü hareketçisi mesafesi olarak aldandırılan bu yöntem hem derinlik hem de iskelet bilgisini kullanarak önceki yöntemin dezavantajı olan siyah bileklik olmadan da el bölgesini bölütleyebilmiştir.…”

Section: Introductionunclassified

“…Bu durum gen için geçerlidir [1,7]. Eğer kameraya eld nesne bulunması durumunda, hem renk görüntülerini birlikte kullanabilecek algoritması kullanılabilir [5,6]. Böylec histogramında gölgelerden sonra 0 gri se yakın nesne el olacaktır.…”

Section: Introductionunclassified

Hand gesture recognition from kinect depth images

Yeloglu

Akbulut

Budak

et al. 2015

2015 23nd Signal Processing and Communications Applications Conference (SIU)

View full text Add to dashboard Cite

Özetçe-Bu çalışmada derinlik imgelerine dayalı bir el hareketleri sınıflandırma yöntemi önerilmiştir. Yöntem; eşikleme, öznitelik çıkarma, öznitelik seçme ve sınıflandırma aşamalarından oluşmaktadır. Derinlik imgelerinden elin bölütlenmesi için aralık eşikleme, öznitelik çıkarımı için eğrilik ölçek uzayı, öznitelik indirgeme için sıralı öznitelik seçme ve sınıflandırma için de K-En Yakın Komşu yöntemi kullanılmıştır. Önerilen yöntemin başarım değerlendirmesi için 1000 örnekten oluşan bir veri seti kullanılmıştır. Deneysel çalışmalar önerilen yöntemin 0'dan 9'a kadar olan el işaretlerini %98.33'lük bir doğrulukla sınıflandırabildiğini göstermiştir. Bu oran karşılaştırılan yöntemden %4 civarında daha iyidir. Anahtar Kelimeler -Derinlik imgeleri, El hareketi, Eğrilik ölçek uzayı, sınıflandırma.Abstract-In this study, hand gesture classification method based on depth images is proposed. The proposed method is composed of thresholding, feature extraction, feature selection and classification stages. Hand segmentation on the depth images is carried out based on interval thresholding, curvature scale space is used for feature extraction, sequential feature selection is considered for feature selection and K-Nearest Neighbor method is used for classification. The performance evaluation of the proposed method is tested on 1000 sampled dataset. Experimental works show that the hand gestures which indicate from 0 to 9 can be recognized with 98.33 % accuracy. This accuracy rate is about 4% better than the compared method.

show abstract

Superpixel-Based Hand Gesture Recognition With Kinect Depth Camera

Cited by 284 publications

References 33 publications

Fast and robust dynamic hand gesture recognition via key frames extraction and feature fusion

Fast and robust dynamic hand gesture recognition via key frames extraction and feature fusion

BASS: Boundary-Aware Superpixel Segmentation

Hand gesture recognition from kinect depth images

Contact Info

Product

Resources

About