Player detection method based on scale attention and scale equalization algorithm

Review Computer Science, Information Systems

Review and evaluation of player detection methods in field sports Comparing conventional and deep learning based methods

Melike Sah et al.

Summary: Accurate player detection is crucial for analyzing field sport videos. Various methods, including conventional ones, convolutional neural networks, and deep learning techniques, have been proposed for player detection. This study aims to compare the performances of these methods on field hockey datasets under the same evaluation settings.

MULTIMEDIA TOOLS AND APPLICATIONS (2023)

Add to Collection

Article Computer Science, Information Systems

Use of deep learning in soccer videos analysis: survey

Sara Akan et al.

Summary: The demand for video analysis has increased rapidly in the past decade, especially in fields such as medical diagnosis, security surveillance, robotics, and sports. This study focuses on deep learning-based techniques for analyzing football videos, aiming to show the most recent research in the field instead of comparing current methodologies. The study investigates the challenges and application groups of soccer video analysis, as well as reviewing the performance, advantages, and disadvantages of deep learning-based methods used in soccer videos and discussing future potential in this area.

MULTIMEDIA SYSTEMS (2023)

Add to Collection

Article Computer Science, Artificial Intelligence

Graph-based few-shot learning with transformed feature propagation and optimal class allocation

Ruiheng Zhang et al.

Summary: The paper proposes a method to transform features extracted by a pre-trained self-supervised feature extractor into a Gaussian-like distribution to reduce feature distribution mis-match, and to solve incorrect class allocation through computing an optimal class allocation matrix.

NEUROCOMPUTING (2022)

Add to Collection

Article Mathematical & Computational Biology

Deep Learning-Based Football Player Detection in Videos

Tianyi Wang et al.

Summary: This study proposes a deep convolutional neural network-based football video analysis algorithm for detecting and tracking football players in real time. The algorithm utilizes five convolution blocks to extract feature maps of players with different spatial resolutions. By combining features from different levels and using weighted parameters, it improves detection accuracy and adapts to input images of varying resolutions and qualities. The experimental results confirm the effectiveness of the algorithm.

COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE (2022)

Add to Collection

Article Computer Science, Information Systems

Directed searching optimized mean-exposure based sub-image histogram equalization for grayscale image enhancement

Upendra Kumar Acharya et al.

Summary: This paper introduces a novel enhancement technique for low light grayscale images, which improves visual quality and information contents through optimized histogram equalization. By clipping and dividing the histogram, the method prevents over enhancement and preserves natural appearance.

MULTIMEDIA TOOLS AND APPLICATIONS (2021)

Add to Collection

Proceedings Paper Computer Science, Artificial Intelligence

Construction worker hardhat-wearing detection based on an improved BiFPN

Chenyang Zhang et al.

Summary: Work in construction sites is considered high-risk, and safety is crucial. A one-stage object detection method based on convolutional neural network was proposed, which effectively identifies small-scale hardhats and achieves higher mAP through an improved approach.

2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR) (2021)

Add to Collection

Article Engineering, Electrical & Electronic