Cycle Training with Semi-Supervised Domain Adaptation: Bridging Accuracy and Efficiency for Real-Time Mobile Scene Detection
This addresses the problem of efficient AI deployment on resource-limited mobile devices for scene detection, representing a strong specific gain.
The paper tackles the challenge of balancing accuracy and computational efficiency for real-time mobile scene detection by proposing a Cycle Training framework with Semi-Supervised Domain Adaptation, achieving 94.00% Top-1 accuracy and 1.61ms inference time on CPU.
Nowadays, smartphones are ubiquitous, and almost everyone owns one. At the same time, the rapid development of AI has spurred extensive research on applying deep learning techniques to image classification. However, due to the limited resources available on mobile devices, significant challenges remain in balancing accuracy with computational efficiency. In this paper, we propose a novel training framework called Cycle Training, which adopts a three-stage training process that alternates between exploration and stabilization phases to optimize model performance. Additionally, we incorporate Semi-Supervised Domain Adaptation (SSDA) to leverage the power of large models and unlabeled data, thereby effectively expanding the training dataset. Comprehensive experiments on the CamSSD dataset for mobile scene detection demonstrate that our framework not only significantly improves classification accuracy but also ensures real-time inference efficiency. Specifically, our method achieves a 94.00% in Top-1 accuracy and a 99.17% in Top-3 accuracy and runs inference in just 1.61ms using CPU, demonstrating its suitability for real-world mobile deployment.