Kinematic Spline Curves: A temporal invariant descriptor for fast action recognition

Computer Methods and Programs in Biomedicine

Shabayek

et al. 2019

Self Cite

Background and Objective: With the increase in the number of stroke survivors, there is an urgent need for designing appropriate home-based rehabilitation tools to reduce health-care costs. The objective is to empower the rehabilitation of post-stroke patients at the comfort of their homes by supporting them while exercising without the physical presence of the therapist. Methods: A novel low-cost home-based training system is introduced. This system is designed as a composition of two linked applications: one for the therapist and another one for the patient. The therapist prescribes personalized exercises remotely, monitors the home-based training and re-adapts the exercises if required. On the other side, the patient loads the prescribed exercises, trains the prescribed exercise while being guided by color-based visual feedback and gets updates about the exercise performance. To achieve that, our system provides three main functionalities, namely: 1) Feedback proposals guiding a personalized exercise session, 2) Posture monitoring optimizing the effectiveness of the session, 3) Assessment of the quality of the motion. Results: The proposed system is evaluated on 10 healthy participants without any previous contact with the system. To analyze the impact of the feedback proposals, we carried out two different experimental sessions: without and with feedback proposals. The obtained results give preliminary assessments about the interest of using such feedback. Conclusions: Obtained results on 10 healthy participants are promising. This encourages to test the system in a realistic clinical context for the rehabilitation of stroke survivors.

“…• Exercise accuracy γ. It quantifies the quality of the patient motion based on the temporal alignment proposed in [33].…”

Section: Measurements Analysismentioning

confidence: 99%

Section: Exercise Accuracymentioning

confidence: 99%

Section: Exercise Accuracymentioning

confidence: 99%

See 1 more Smart Citation

Home self-training: Visual feedback for assisting physical activity for stroke survivors

Baptista

Computer Methods and Programs in Biomedicine

Shabayek

et al. 2019

Self Cite

“…To overcome the anthropometric variability, we propose to follow the same protocol presented in (Ghorbel et al, 2018). All joint positions are normalized except the hip joint position which is assumed to be the root and is therefore unchanged (p norm hip = p hip ).…”

Section: Skeleton Normalizationmentioning

confidence: 99%

An extension of kernel learning methods using a modified Log-Euclidean distance for fast and accurate skeleton-based Human Action Recognition

Computer Vision and Image Understanding

Boonært

Boutteau

et al. 2018

Self Cite

In this article, we introduce a fast, accurate and invariant method for RGB-D based human action recognition using a Hierarchical Kinematic Covariance (HKC) descriptor.Recently, non singular covariance matrices of pattern features which are elements of the space of Symmetric Definite Positive (SPD) matrices, have been proven to be very efficient descriptors in the field of pattern recognition.However, in the case of action recognition, singular covariance matrices cannot be avoided because the dimension of features could be higher than the number of samples. Such covariance matrices (non singular and singular) belong to the space of Symmetric Positive semi-Definite (SPsD) matrices.Thus, in order to classify actions, we propose to adapt kernel methods such as Support Vector Machines (SVM) and Multiple Kernel Learning (MKL) to the space of SPsD matrices by using a perturbed Log-Euclidean distance (Arsigny et al., 2006). The mathematical validity of this perturbed distance (called Modified Log-Euclidean distance) for SPsD is therefore studied.The offline experiments are conducted on three challenging benchmarks, namely MSRAction3D, UTKinect and Multiview3D datasets. A fair comparison demonstrates that our approach competes with state-of-the-art methods in terms of accuracy and computational latency. Finally, our method is extended to an online scenario and experiments on MSRC12 prove the efficiency of this extension. (Enjie Ghorbel) can be found in (Poppe, 2010;Weinland et al., 2011). Unfortunately, these methods suffer from some limitations: their performance is negatively affected by occlusions, view-point variation, illumination changes and body segmentation.With the availability of low-cost RGB-D cameras, a renewed interest for action recognition has been observed. Additionally to the classical RGB images, this kind of camera also provides depth images. Furthermore, the recent algorithm proposed by Shotton et al. (2013) allows the real-time human skeleton extraction from depth maps. Thus, new methods (Rahmani et al., 2016;Amor et al., 2016;Brun et al., 2016;Liu et al., 2016) have been pro-

“…The introduction of RGB-D cameras played an important role in enhancing view-invariant action recognition. In fact, these low-cost sensors provide in real-time relatively accurate 3D skeletons that have boosted the design of viewinvariant approaches [2,3,4,5,6,7]. We distinguish two different ways of addressing the issue of viewpoint variability using skeletons provided by RGB-D sensors.…”

Section: Introductionmentioning

confidence: 99%

View-invariant Action Recognition from RGB Data via 3D Pose Estimation

Baptista

ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Papadopoulos

et al. 2019

Self Cite

In this paper, we propose a novel view-invariant action recognition method using a single monocular RGB camera. Viewinvariance remains a very challenging topic in 2D action recognition due to the lack of 3D information in RGB images. Most successful approaches make use of the concept of knowledge transfer by projecting 3D synthetic data to multiple viewpoints. Instead of relying on knowledge transfer, we propose to augment the RGB data by a third dimension by means of 3D skeleton estimation from 2D images using a CNN-based pose estimator. In order to ensure viewinvariance, a pre-processing for alignment is applied followed by data expansion as a way for denoising. Finally, a Long-Short Term Memory (LSTM) architecture is used to model the temporal dependency between skeletons. The proposed network is trained to directly recognize actions from aligned 3D skeletons. The experiments performed on the challenging Northwestern-UCLA dataset show the superiority of our approach as compared to state-of-the-art ones.