DeepSketch 2: Deep convolutional neural networks for partial sketch recognition

Freehand sketches are a simple and powerful tool for communication. They are easily recognized across cultures and suitable for various applications. In this paper, we propose a new approach for partial sketch recognition. This could be used to design applications using real-time sketch recognition. We use deep convolutional neural networks (ConvNets), state-of-the-art in the field of sketch recognition. To the best of our knowledge, it is the first ConvNet for partial sketch classification. Our first aim is to build a ConvNet capable of recognizing partial sketches without compromising the accuracy reached for complete sketch recognition. Therefore, we evaluate different approaches and propose an efficient way for partial sketch recognition. Our second aim is improving complete sketch recognition using information about sketching progression. We obtained a ConvNet that outperforms state-of-the-art results in the TU-Berlin sketch benchmark. We reached an accuracy of 77.69%.