Motion Recognition Based on Deep Learning and Human Joint Points

Jun-ping, Wang

doi:10.1155/2022/1826951

Cited by 3 publications

(2 citation statements)

References 20 publications

(15 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cosine similarity is a widely used algorithm for calculating the similarity between two vectors. It is commonly used in text analysis, image recognition, and identification of medicinal plants (Wang, 2022; Zakeri & Saradjian, 2022). The cosine angle algorithm was used to calculate the similarity between the samples (Figure 4b).…”

Section: Resultsmentioning

confidence: 99%

Diversity in Pyracantha fortuneana fruits maturity stages enables discrepancy in the phenolic compounds, antioxidant activity, and tyrosinase inhibitory activity

Wang,

Shao,

Yang

et al. 2024

Journal of Food Science

View full text Add to dashboard Cite

Pyracantha fortuneana (P. fortuneana) fruit is a wild fruit that is popular because of its delicious taste and numerous nutrients, and phenolic compounds are considered to be the main bioactive components in P. fortuneana fruits. However, the relationship between phenolic compounds and their antioxidant and tyrosinase (TYR) inhibitory activities during the ripening process is still unclear. The study compared the influence of the five developmental stages on the accumulation of phenolic compounds, antioxidant activity, and TYR inhibitory activity in the fruits of P. fortuneana. The compounds were identified by offline two‐dimensional liquid chromatography‐electrochemical detection (2D‐LC‐ECD) combined with liquid chromatography–tandem mass spectrometry, and the main active ingredients were quantified. The results showed that stage II had higher total phenolic and flavonoid content, as well as higher antioxidant and TYR inhibitory activity, but the total anthocyanin content was lowest at this stage. A total of 30 compounds were identified by 2D‐LC‐ECD. Orthogonal partial least squares discriminant analysis screened out six major potential markers, including phenolic acids, procyanidins, and flavonoids. In addition, it was found that caffeoylquinic acids, procyanidins, and flavonoids were higher in stage II than in stages I, III, IV, and V, whereas anthocyanins accumulated gradually from stages III to V. Therefore, this study suggests that the changes in antioxidant and TYR inhibitory activities of P. fortuneana during the five developmental stages may be due to the transformation of procyanidins, caffeoylquinic acids, and phenolic glycosides into other forms during the fruit maturation process.Practical Application: Differences in chemical constituents, antioxidant, and tyrosinase inhibitory activities in fruit maturity stages of P. fortuneana were elucidated to provide reference for rational harvesting and utilization of the fruits and their bioactive components. These findings are expected to provide a comprehensive assessment of the bioactive profile and guide the food industrial production.

show abstract

Section: Resultsmentioning

confidence: 99%

Diversity in Pyracantha fortuneana fruits maturity stages enables discrepancy in the phenolic compounds, antioxidant activity, and tyrosinase inhibitory activity

Wang,

Shao,

Yang

et al. 2024

Journal of Food Science

View full text Add to dashboard Cite

show abstract

“…The most commonly used sensor is the IMU sensor, which achieves an exercise classification accuracy of 95% or higher [ 30 , 31 , 32 , 33 ]. Joint position data is also used for exercise type recognition [ 34 ]. These sensors are used for not only exercise type recognition but also for counting the repetitions of the exercise [ 32 , 33 ].…”

Section: Introductionmentioning

confidence: 99%

Estimation of One-Repetition Maximum, Type, and Repetition of Resistance Band Exercise Using RGB Camera and Inertial Measurement Unit Sensors

Hwang

Shim

Choi

et al. 2023

Sensors

View full text Add to dashboard Cite

Resistance bands are widely used nowadays to enhance muscle strength due to their high portability, but the relationship between resistance band workouts and conventional dumbbell weight training is still unclear. Thus, this study suggests a convolutional neural network model that identifies the type of band workout and counts the number of repetitions and a regression model that deduces the band force that corresponds to the one-repetition maximum. Thirty subjects performed five different exercises using resistance bands and dumbbells. Joint movements during each exercise were collected using a camera and an inertial measurement unit. By using different types of input data, several models were created and compared. As a result, the accuracy of the convolutional neural network model using inertial measurement units and joint position is 98.83%. The mean absolute error of the repetition counting algorithm ranges from 0.88 (seated row) to 3.21(overhead triceps extension). Lastly, the values of adjusted r-square for the 5 exercises are 0.8415 (chest press), 0.9202 (shoulder press), 0.8429 (seated row), 0.8778 (biceps curl), and 0.9232 (overhead triceps extension). In conclusion, the model using 10-channel inertial measurement unit data and joint position data has the best accuracy. However, the model needs to improve the inaccuracies resulting from non-linear movements and one-time performance.

show abstract

Robust Keypoint Detection and Matching on Fisheye Images by Self-Supervised Learning

Tian

Cai

Wen

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Accurate image feature point detection and matching are essential to computer vision tasks such as panoramic image stitching and 3D reconstruction. However, ordinary feature point approaches cannot be directly applied to fisheye images due to their large distortion, which makes the ordinary camera model unable to adapt. To address such a problem, this paper proposes a self-supervised learning method for feature point detection and matching on fisheye images. This method utilizes a Siamese network to automatically learn the correspondence of feature points across transformed image pairs to avoid high annotation costs. Due to the scarcity of the fisheye image dataset, a two-stage viewpoint transform pipeline is also adopted for image augmentation to increase the data variety. Furthermore, this method adopts both deformable convolution and contrastive learning loss to improve the feature extraction and description of distorted image regions. Compared with traditional feature point detectors and matchers, this method has been demonstrated with superior performance on fisheye images.

show abstract

Motion Recognition Based on Deep Learning and Human Joint Points

Cited by 3 publications

References 20 publications

Diversity in Pyracantha fortuneana fruits maturity stages enables discrepancy in the phenolic compounds, antioxidant activity, and tyrosinase inhibitory activity

Diversity in Pyracantha fortuneana fruits maturity stages enables discrepancy in the phenolic compounds, antioxidant activity, and tyrosinase inhibitory activity

Estimation of One-Repetition Maximum, Type, and Repetition of Resistance Band Exercise Using RGB Camera and Inertial Measurement Unit Sensors

Robust Keypoint Detection and Matching on Fisheye Images by Self-Supervised Learning

Contact Info

Product

Resources

About