Sketch Based Model-Like Standing Style Recommendation.

Various mobile devices with high-quality cameras are very popular in human daily life. Appropriate directions about the standing postures can greatly improve the user experience while taking photos. In this paper, we propose a method to recommend custom model-like standing style based on model sketches. We first translate the real images of splendid models into sketches by fast person detection and model sketching. The generated sketches are represented by the deep feature vectors. We design an iterative detection approach to finding the most representative model sketches. For a user-input image, the model-like standing styles are recommended in form of the sketch. Besides, we introduce a novel method to score the standing posture of the input image through multi-Gaussian functions. Finally, the experimental results on the Model Standing Style (MSS) dataset demonstrate the effectiveness of the proposed approach.