One-shot learning gesture recognition from RGB-D data using bag of features

By Jun Wan, Q. Ruan, Shuang Deng. 2013

120 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition, Human-Computer Interaction

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

In Computer Vision and Pattern Recognition

In Human-Computer Interaction

Citations in the corpus

RGB-D-based Human Motion Recognition with Deep Learning: A Survey (2017, methodology): “Wang et al. (2016g) leveraged semantic cues in video by using a twostream semantic region-based CNNs (SR-CNNs) to incorporate human/ object detection results into the framework.”
One-shot-learning gesture recognition using HOG-HOF features (2013, methodology): “Bag-of-features models provide comparable (Wan et al., 2013) or slightly worse (HernándezVela et al., 2012) results than ours.”
Domain-Adaptive Discriminative One-Shot Learning of Gestures (2014, background): “As a result, several recent works have attempted to learn gestures at the other extreme – from single training examples using one-shot learning [16,17,19,20,22,24,33].”
ChaLearn Looking at People RGB-D Isolated and Continuous Datasets for Gesture Recognition (2016, methodology): “The predicted temporal segmentations are obtained by the dynamic time warping (DTW) algorithm [24].”
Small Sample Learning in Big Data Era (2018, methodology): “Wan et al. (2013) proposed a new spatio-temporal feature representation (3D EMoSIFT) by fusing RGB-D data, which was invariant to scale and rotation, and then used nearest neighbor classifier for one-shot learning gesture recognition.”
3D SMoSIFT: three-dimensional sparse motion scale invariant feature transform for activity recognition from RGB-D videos (2014, methodology): “Although it is simple to calculate the depth velocity, the poor performance in 3D motion space is proved by Wan et al.(14) Therefore, we modified 3D motion space as shown in Fig.”
A Unified Framework for Multi-Modal Isolated Gesture Recognition (2018, unknown): “Manyhand-crafted features have been proposed for video analysis in the area of gesture recognition [Starner et al. 1998; Wang et al. 2006; Dardas and Georganas 2011; Trindade et al. 2012; Wan et al. 2013].”
Challenges in multimodal gesture recognition (2016, methodology): “The proposed method did not do as well as the previous methods (Konecny and Hagara, 2014; Wan et al., 2013; Jiang et al., 2015) on the CDG 2011 dataset, but nonetheless it contributes novel ideas for transfer learning, that can be useful when the number of training examples per class is limited.”
Multi-Modality Fusion based on Consensus-Voting and 3D Convolution for Isolated Gesture Recognition (2016, background): “Many hand-crafted features have been proposed for video analysis in the area of gesture recognition [28, 39, 2, 31, 35].”
ChaLearn looking at people: A review of events and resources (2017, background): “Topic at Journal of Machine Learning Research [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30] 2016 TPAMI HuPBA: We organized a Special Issue on Multi-modal Human Pose Recovery and Behavior Analysis (HuPBA) at IEEE Transactions on Pattern Ana”
Learning to Learn: Model Regression Networks for Easy Small Sample Learning (2016, background): “In particular, inter-class transfer [40] and crossgeneralization [7] are achieved by discovering shared feature representations: 1) captured by linear or nonlinear feature transformations [14, 63, 85, 31, 1, 94, 48], 2) obtained by feature selection [59, 60, 27] or regularization [37], 3) described ”
Large-scale gesture recognition with a fusion of RGB-D data based on the C3D model (2016, background): “extend the scale invariant feature transform (SIFT) and propose 3D enhanced motion SIFT (3D EMoSIFT) [11] and 3D Sparse Motion SIFT (3D SMoSIFT) [12] to extract features with the fusion of RGBD data.”
Explore Efficient Local Features from RGB-D Data for One-Shot Learning Gesture Recognition (2016, methodology): “[67] 74.8 Ming et al. 2012 [50] 75.9 Li and Ling et al. 2013 [68] 78.1 Wan et al. 2013 [9] 86.0 Wan et al. 2014 [20] 93.2 Ours (MFSK+BoVW) 95.7”
Depth Pooling Based Large-Scale 3-D Action Recognition With Convolutional Neural Networks (2018, methodology): “used method employs dynamic time warping (DTW) to determine the delimiting frames of individual actions [27]–[29].”
Results and Analysis of the ChaLearn Gesture Challenge 2012 (2012, unknown): “The next best ranked participants (who won third place in round 2), the Joewan team [15], used a slightly different approach.”
Multi-layered gesture recognition with Kinect (2015, methodology): “Wan et al. (2013, 2014b) extend SIFT to spatio-temporal domain and propose 3D EMoSIFT and 3D SMoSIFT to extract features from RGB and depth images, which are invariant to scale and rotation, and have more compact and richer visual representations.”