Sports Classification, a cutting-edge computer vision task, harnesses the power of MobileNet transfer learning to accurately classify images across 100 sports categories. Achieving outstanding accuracy, with a train accuracy of 97.9% and validation accuracy of 98.88% post fine-tuning, it stands as a robust solution for sports image recognition.
-
MobileNet Transfer Learning: Leverages the efficiency of MobileNet architecture for effective model training.
-
Diverse Sports Categories: Recognizes and classifies images across a wide spectrum of 100 sports categories.
-
High Accuracy: Boasts impressive performance with a train accuracy of 97.9% and a validation accuracy of 98.88%.
-
Computer Vision Task: Tailored for the nuances of computer vision, ensuring accurate and efficient image classification.
-
Easy Integration: Seamlessly integrates into applications and workflows, supporting various platforms.
-
Data Preparation: Organize images into categories representing the diverse world of sports.
-
Model Training: Utilize MobileNet transfer learning for swift and accurate training, achieving exceptional accuracy in this computer vision task.
-
Inference: Deploy the model for real-time sports image classification across 100 different categories.
We welcome contributions and feedback to enhance the Sports Classifier's capabilities. Let's build an advanced and accurate sports image recognition system together.
Dataset gotten from kaggle '100 Sports Image Classification' competition. https://www.kaggle.com/datasets/gpiosenka/sports-classification