Recent learning-based approaches show promising performance improvement for the scene text removal task but usually leave several remnants of text and provide visually unpleasant results. In this work, a novel end-to-end framework is proposed based on accurate text stroke detection. Specifically, the text removal problem is decoupled into text stroke detection and stroke removal; we design separate networks to solve these two subproblems, the latter being a generative network. These two networks are combined as a processing unit, which is cascaded to obtain our final model for text removal. Experimental results demonstrate that the proposed method substantially outperforms the state-of-the-art for locating and erasing scene text. A new large-scale real-world dataset with 12,120 images has been constructed and is being made available to facilitate research, as current publicly available datasets are mainly synthetic so cannot properly measure the performance of different methods.

Existing physical cloth simulators suffer from expensive computation and difficulties in tuning mechanical parameters to get desired wrinkling behaviors. Data-driven methods provide an alternative solution. They typically synthesize cloth animation at a much lower computational cost, and also create wrinkling effects that are similar to the training data. In this paper we propose a deep learning based method for synthesizing cloth animation with high resolution meshes. To do this we first create a dataset for training: a pair of low and high resolution meshes are simulated and their motions are synchronized. As a result the two meshes exhibit similar large-scale deformation but different small wrinkles. Each simulated mesh pair is then converted into a pair of low- and high-resolution “images” (a 2D array of samples), with each image pixel being interpreted as any of three descriptors: the displacement, the normal and the velocity. With these image pairs, we design a multi-feature super-resolution (MFSR) network that jointly trains an upsampling synthesizer for the three descriptors. The MFSR architecture consists of shared and task-specific layers to learn multi-level features when super-resolving three descriptors simultaneously. Frame-to-frame consistency is well maintained thanks to the proposed kinematics-based loss function. Our method achieves realistic results at high frame rates: 12–14 times faster than traditional physical simulation. We demonstrate the performance of our method with various experimental scenes, including a dressed character with sophisticated collisions.