- |
-
Converge intra-class and Diverge inter-class features for CNN-based Event Detection in football videos
Oral Presentation
Authors
1Department of Computer Engineering, North Tehran Branch, Islamic Azad University, Tehran, Iran
2Faculty of Computer Engineering, Air University, Tehran, Iran
3Faculty of Electrical and Computer Engineering, Quchan University of Technology, Quchan, Iran
Abstract
Football event detection in videos is very
challenging, but challenges on the Penalty and the Free-kick,
which have common visual elements, are severe and critical. The
existence of common elements between two events causes the
extraction of common and ineffective features in recognizing
these two events. As a result, the error of recognizing and
separating these two events is more than other events. In this
paper, we present a new method for filtering the input data to
converge the intra-class features and diverge the inter-class
features to increase the classification accuracy. For this purpose,
we have evaluated images for the Penalty and the Free-kick
classes with the criterion of structural similarity. Based on the
results, inappropriate images have been ignored according to the
average value and standard deviation of each class of data. This
filtration leads to ignore of ineffective and common features in
the learning process. The results of the proposed method indicate
an improvement in the accuracy of distinguishing between two
Penalty and Free-kick events using a deep neural network and
filtered training images compared to the deep neural network
using all training images.
challenging, but challenges on the Penalty and the Free-kick,
which have common visual elements, are severe and critical. The
existence of common elements between two events causes the
extraction of common and ineffective features in recognizing
these two events. As a result, the error of recognizing and
separating these two events is more than other events. In this
paper, we present a new method for filtering the input data to
converge the intra-class features and diverge the inter-class
features to increase the classification accuracy. For this purpose,
we have evaluated images for the Penalty and the Free-kick
classes with the criterion of structural similarity. Based on the
results, inappropriate images have been ignored according to the
average value and standard deviation of each class of data. This
filtration leads to ignore of ineffective and common features in
the learning process. The results of the proposed method indicate
an improvement in the accuracy of distinguishing between two
Penalty and Free-kick events using a deep neural network and
filtered training images compared to the deep neural network
using all training images.
Keywords
convolutional neural networks; intra-class features; inter-class features; VGGnet; ResNet; Common features
Proceeding Title [Persian]
Converge intra-class and Diverge inter-class features for CNN-based Event Detection in football videos
Authors [Persian]
Abstract [Persian]
Football event detection in videos is very
challenging, but challenges on the Penalty and the Free-kick,
which have common visual elements, are severe and critical. The
existence of common elements between two events causes the
extraction of common and ineffective features in recognizing
these two events. As a result, the error of recognizing and
separating these two events is more than other events. In this
paper, we present a new method for filtering the input data to
converge the intra-class features and diverge the inter-class
features to increase the classification accuracy. For this purpose,
we have evaluated images for the Penalty and the Free-kick
classes with the criterion of structural similarity. Based on the
results, inappropriate images have been ignored according to the
average value and standard deviation of each class of data. This
filtration leads to ignore of ineffective and common features in
the learning process. The results of the proposed method indicate
an improvement in the accuracy of distinguishing between two
Penalty and Free-kick events using a deep neural network and
filtered training images compared to the deep neural network
using all training images.
challenging, but challenges on the Penalty and the Free-kick,
which have common visual elements, are severe and critical. The
existence of common elements between two events causes the
extraction of common and ineffective features in recognizing
these two events. As a result, the error of recognizing and
separating these two events is more than other events. In this
paper, we present a new method for filtering the input data to
converge the intra-class features and diverge the inter-class
features to increase the classification accuracy. For this purpose,
we have evaluated images for the Penalty and the Free-kick
classes with the criterion of structural similarity. Based on the
results, inappropriate images have been ignored according to the
average value and standard deviation of each class of data. This
filtration leads to ignore of ineffective and common features in
the learning process. The results of the proposed method indicate
an improvement in the accuracy of distinguishing between two
Penalty and Free-kick events using a deep neural network and
filtered training images compared to the deep neural network
using all training images.
Keywords [Persian]
convolutional neural networks، intra-class features، inter-class features، VGGnet، ResNet، Common features