LinkTo-Anime: A 2D Animation Optical Flow Dataset from 3D Model Rendering
Not on the map. Why?
Abstract (arXiv)
Existing optical flow datasets focus primarily on real-world simulation or synthetic human motion, but few are tailored to Celluloid(cel) anime character motion: a domain with unique visual and motion characteristics. To bridge this gap and facilitate research in optical flow estimation and downstream tasks such as anime video generation and line drawing colorization, we introduce LinkTo-Anime, the first high-quality dataset specifically designed for cel anime character motion generated with 3D model rendering. LinkTo-Anime provides rich annotations including forward and backward optical flow, occlusion masks, and Mixamo Skeleton. The dataset comprises 395 video sequences, totally 24,230 training frames, 720 validation frames, and 4,320 test frames. Furthermore, a comprehensive benchmark is constructed with various optical flow estimation methods to analyze the shortcomings and limitations across multiple datasets.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 27 of its 37 references that are in the map. First posted on arXiv 3 Jun 2025.
Builds on
27 of 37 references.
References · 27
- 2017PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volumecited by 968
- 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 881
- 2016Virtual Worlds as Proxy for Multi-Object Tracking Analysiscited by 341
- 2021GMFlow: Learning Optical Flow via Global Matchingcited by 327
- 2020Softmax Splatting for Video Frame Interpolationcited by 279
- 2022FlowFormer: A Transformer Architecture for Optical Flowcited by 275
- 2021Learning to Estimate Hidden Motions with Global Motion Aggregationcited by 224
- 2020Real-Time Intermediate Flow Estimation for Video Frame Interpolationcited by 213
- 2023Conditional Image-to-Video Generation with Latent Flow Diffusion Modelscited by 181
- 2019Quadratic video interpolationcited by 145
- 2014A Tutorial on Principal Component Analysiscited by 141
- 2020MaskFlownet: Asymmetric Feature Matching with Learnable Occlusion Maskcited by 117
15 more
- 2023FlowFormer++: Masked Cost Volume Autoencoding for Pretraining Optical Flow Estimationcited by 114
- 2024MemFlow: Optical Flow Estimation and Prediction with Memorycited by 55
- 2022A Unified Pyramid Recurrent Network for Video Frame Interpolationcited by 39
- 2024Anim-Director: A Large Multimodal Model Powered Agent for Controllable Animation Video Generationcited by 37
- 2022Disentangling Architecture and Training for Optical Flowcited by 32
- 2024LVCD: Reference-based Lineart Video Colorization with Diffusion Modelscited by 27
- 2021The Animation Transformer: Visual Correspondence via Segment Matchingcited by 20
- 2023PAniC-3D: Stylized Single-view 3D Reconstruction from Portraits of Anime Characterscited by 17
- 2023DistractFlow: Improving Optical Flow Estimation via Realistic Distractions and Pseudo-Labelingcited by 14
- 2024Learning Inclusion Matching for Animation Paint Bucket Colorizationcited by 14
- 2022AnimeRun: 2D Animation Visual Correspondence from Open Source 3D Moviescited by 11
- 2023Deep Geometrized Cartoon Line Inbetweeningcited by 9
- 2023CogCartoon: Towards Practical Story Visualizationcited by 8
- 2022SketchBetween: Video-to-Video Synthesis for Sprite Animation via Sketchescited by 6
- 2024Paint Bucket Colorization Using Anime Character Color Design Sheetscited by 3
Built on by
Citing papers per year: 2026: 2.
2 papers.