Fast and Accurate Quantized Camera Scene Detection on Smartphones, Mobile AI 2021 Challenge: Report

Camera scene detection is among the most popular computer vision problem on\nsmartphones. While many custom solutions were developed for this task by phone\nvendors, none of the designed models were available publicly up until now. To\naddress this problem, we introduce the first Mobile AI challenge, where the\ntarget is to develop quantized deep learning-based camera scene classification\nsolutions that can demonstrate a real-time performance on smartphones and IoT\nplatforms. For this, the participants were provided with a large-scale CamSDD\ndataset consisting of more than 11K images belonging to the 30 most important\nscene categories. The runtime of all models was evaluated on the popular Apple\nBionic A11 platform that can be found in many iOS devices. The proposed\nsolutions are fully compatible with all major mobile AI accelerators and can\ndemonstrate more than 100-200 FPS on the majority of recent smartphone\nplatforms while achieving a top-3 accuracy of more than 98%. A detailed\ndescription of all models developed in the challenge is provided in this paper.\n