Creative Flow+ Dataset
Abstract
We present the Creative Flow+ Dataset, the first diverse multi-style artistic video dataset richly labeled with per-pixel optical flow, occlusions, correspondences, segmentation labels, normals, and depth. Our dataset includes 3000 animated sequences rendered using styles randomly selected from 40 textured line styles and 38 shading styles, spanning the range between flat cartoon fill and wildly sketchy styles. Our dataset includes 124K+ train set frames and 10K test set frames, far surpassing the largest available optical flow datasets in size. While modern techniques for tasks such as optical flow estimation achieve impressive performance on realistic images and video, today there is no way to gauge their performance on stylized content. Creative Flow+ poses a new challenge to generalize real-world Computer Vision to messy stylized content. We show that learning-based optical flow methods fail to generalize to this data and struggle to compete with classical approaches, and invite new research into this area. Our new optical flow benchmark will be hosted online, with a public leaderboard to track progress. We further plan to release the complete dataset creation pipeline, allowing the community to generate and stylize their own data on demand.