Research map / Video / Video generation
FlowStyler: Artistic Video Stylization via Transformation Fields Transports
Yuning Gong, Jiaming Chen, Xiaohua Ren, Yuanjun Liao, Yanci Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use FlowStyler instead of CAP-VSTNet | video stylization · DAVIS | warping error 1-frame: 4.1 -> 2.5 | Improved |
| use FlowStyler instead of AnyV2V | video stylization · DAVIS | DINO score: 0.61 -> 0.72 | Improved |
| use FlowStyler instead of CAP-VSTNet | video stylization · DAVIS | DINO score: 0.58 -> 0.72 | Improved |
| use FlowStyler instead of TokenFlow | video stylization · DAVIS | CLIP score: 0.65 -> 0.67 | No clear effect |
| use FlowStyler instead of TokenFlow for memory | video stylization · DAVIS | peak GPU memory 100 frames (GB): 64 -> 3.6 | Improved |
| use FlowStyler instead of CAP-VSTNet for memory | video stylization · DAVIS | peak GPU memory 100 frames (GB): 0.44 -> 3.6 | Hurt |
| use orthogonality-regularized color field instead of additive field (ablation) | FlowStyler · video stylization | over-stylization (qualitative) | Improved |
| use Lagrangian advection instead of semi-Lagrangian advection (ablation) | FlowStyler · video stylization | noise (qualitative) | Improved |
| add momentum-preserving field optimization (ablation) | FlowStyler · video stylization | vibration artifacts (qualitative) | Improved |
| lower the learning rate instead of preserving momentum (ablation) | FlowStyler · video stylization | blur (qualitative) | Hurt |
| add occlusion-aware temporal lookup (ablation) | FlowStyler · video stylization | trailing artifacts (qualitative) | Improved |
Closest papers by meaning
- Artistic Style Transfer for Videos and Spherical Images (IJCV 2018 · similarity 0.91)
- Stylized Video Generation via Decoupled Data Synthesis and Gated Style Token Injection (ECCV 2026 · similarity 0.90)
- Styl3R: Instant 3D Stylized Reconstruction for Arbitrary Scenes and Styles (NeurIPS 2025 · similarity 0.89)
- StyleMaster: Stylize Your Video with Artistic Generation and Translation (CVPR 2025 · similarity 0.88)
- Morpheus: Text-Driven 3D Gaussian Splat Shape and Color Stylization (CVPR 2025 · similarity 0.88)
- CLIPGaussian: Universal and Multimodal Style Transfer Based on Gaussian Splatting (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
