Fast and Accurate Camera Scene Detection on Smartphones
This addresses the need for fast and accurate scene detection in smartphone cameras, though it is incremental as it builds on existing AI capabilities.
The paper tackles the problem of accurate camera scene detection on smartphones by introducing a new dataset (CamSDD) with over 11K images across 30 categories and proposing an efficient CNN model that achieves 99.5% top-3 accuracy and over 200 FPS on mobile SoCs.
AI-powered automatic camera scene detection mode is nowadays available in nearly any modern smartphone, though the problem of accurate scene prediction has not yet been addressed by the research community. This paper for the first time carefully defines this problem and proposes a novel Camera Scene Detection Dataset (CamSDD) containing more than 11K manually crawled images belonging to 30 different scene categories. We propose an efficient and NPU-friendly CNN model for this task that demonstrates a top-3 accuracy of 99.5% on this dataset and achieves more than 200 FPS on the recent mobile SoCs. An additional in-the-wild evaluation of the obtained solution is performed to analyze its performance and limitation in the real-world scenarios. The dataset and pre-trained models used in this paper are available on the project website.