Exploring feature dimensionality reduction methods for enhancing automatic sport image annotation

Hatem, Yomna; Rady, Sherine

doi:10.1007/s11042-017-5417-z

Cited by 5 publications

(5 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the field of sports, dimensionality reduction techniques have not been used often. In an analysis of the existing literature, only two cases were found, in which they were used for the recovery and annotation of image data sets 35 and to characterize the space-time structure of the kinematics of the throwing action on the direction of the protruding ball. 33…”

Section: Methodsmentioning

confidence: 99%

In-game behaviour analysis of football players using machine learning techniques based on player statistics

García-Aliaga

Marquina

Coterón

et al. 2020

International Journal of Sports Science & Coaching

View full text Add to dashboard Cite

The purpose of this research was to determine the on-field playing positions of a group of football players based on their technical-tactical behaviour using machine learning algorithms. Each player was characterized according to a set of 52 non-spatiotemporal descriptors including offensive, defensive and build-up variables that were computed from OPTA’s on-ball event records of the matches for 18 national leagues between the 2012 and 2019 seasons. To test whether positions could be identified from the statistical performance of the players, the dimensionality reduction techniques were used. To better understand the differences between the player positions, the most discriminatory variables for each group were obtained as a set of rules discovered by RIPPER, a machine learning algorithm. From the combination of both techniques, we obtained useful conclusions to enhance the performance of players and to identify positions on the field. The study demonstrates the suitability and potential of artificial intelligence to characterize players' positions according to their technical-tactical behaviour, providing valuable information to the professionals of this sport.

show abstract

Section: Methodsmentioning

confidence: 99%

In-game behaviour analysis of football players using machine learning techniques based on player statistics

García-Aliaga

Marquina

Coterón

et al. 2020

International Journal of Sports Science & Coaching

View full text Add to dashboard Cite

show abstract

“…Hatem and Rady [15] investigated different feature dimensionality reduction techniques to retrieve and annotate 120 sport images from the Leeds Sports Pose sport dataset. While JSEG algorithm segmented the images, 10 fold cross validation for classification accuracy and performance metrics were evaluated to prove the performance of LSA.…”

Section: Role Of Image Annotationmentioning

confidence: 99%

An Insight on Image Annotation Approaches and their Performances

Et. al.

2021

TURCOMAT

View full text Add to dashboard Cite

Image Annotation (IA) followed by Image Retrieval (IR) plays a significant role in today’s computer vision world. As the manual IA is a tedious and time-consuming process, the automated IA became very predominant in the computer vision applications. IA deals with the assigning of meaningful labels to various objects in the image. The objective of this article is to represent the various IA approaches adopted in the last decade. Observation of the existing IA methods and their performances leads to identify the pitfalls the existing approaches. Few approaches used standard datasets and images downloaded from internet to evaluate the performance of the Image Annotation.

show abstract

“…Beside requiring additional effort and posing a computational overhead, the classification model may also tend to overfit, resulting in poor performance. To avoid such problems in the domain of sports images, in [1], the authors proposed feature dimensionality reduction techniques for the annotation of image datasets. They investigated different techniques, such as Chi-square, information gain, gain ratio, and Latent Semantic Analysis (LSA), and applied them for sports images classification using a SVM classifier.…”

Section: Related Workmentioning

confidence: 99%

“…A huge increase of the amount of non-textual information available in electronic form in mass media requires dealing with it as a major source of content [1]. Digital archiving and multimedia analysis are needed to improve the processing efficiency of this material.…”

Section: Introductionmentioning

confidence: 99%

“…However, in order for CNNs to be able to train a good predictive model with high predictive performance, a large enough dataset with reliable ground truth is needed. While the human annotation is impossible for large image datasets, for smaller datasets the manual process also may not be accurate due to human mistakes and perception subjectivity [1]. In the last years, the lack of adequate datasets is most commonly addressed with the utilization of transfer learning approaches [10].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Classification of Similar Sports Images Using Convolutional Neural Network with Hyper-Parameter Optimization

2020

View full text Add to dashboard Cite

With the exponential growth of the presence of sport in the media, the need for effective classification of sports images has become crucial. The traditional approaches require carefully hand-crafted features, which make them impractical for massive-scale data and less accurate in distinguishing images that are very similar in appearance. As the deep learning methods can automatically extract deep representation of training data and have achieved impressive performance in image classification, our goal was to apply them to automatic classification of very similar sports disciplines. For this purpose, we developed a CNN-TL-DE method for image classification using the fine-tuning of transfer learning for training a convolutional neural network model with the use of hyper-parameter optimization based on differential evolution. Through the automatic optimization of neural network topology and essential training parameters, we significantly improved the classification performance evaluated on a dataset composed from images of four similar sports—American football, rugby, soccer, and field hockey. The analysis of interpretable representation of the trained model additionally revealed interesting insights into how our model perceives images which contributed to a greater confidence in the model prediction. The performed experiments showed our proposed method to be a very competitive image classification method for distinguishing very similar sports and sport situations.

show abstract

Exploring feature dimensionality reduction methods for enhancing automatic sport image annotation

Cited by 5 publications

References 29 publications

In-game behaviour analysis of football players using machine learning techniques based on player statistics

In-game behaviour analysis of football players using machine learning techniques based on player statistics

An Insight on Image Annotation Approaches and their Performances

Classification of Similar Sports Images Using Convolutional Neural Network with Hyper-Parameter Optimization

Contact Info

Product

Resources

About