Research map / Image generation / Editing and style transfer
Scoot: A Perceptual Metric for Facial Sketches
Deng-Ping Fan, ShengChuan Zhang, Yu-Huan Wu, Yun Liu, Ming-Ming Cheng, Bo Ren, Paul L. Rosin, Rongrong Ji
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 Scoot instead of SSIM, FSIM, GMSD, VIF for ranking stability to resizing | Scoot metric · CUFS and CUFSF | MM1: approx 77% and 83% improvement | Improved |
| use Scoot instead of the best classical metric for content capture | Scoot metric · CUFS | MM3: 81.4 -> 95.9 | Improved |
| use Scoot instead of the best classical metric for content capture | Scoot metric · CUFSF | MM3: 97.4 -> 97.5 | No clear effect |
| use Scoot instead of best prior metric for human judgment | Scoot metric · RCUFS | Jud: ~26.3% improvement | Improved |
| add energy feature to contrast (C+E vs C only) (ablation) | Scoot metric · CUFS | MM3: 54.4 -> 95.9 | Improved |
| add contrast feature to energy (C+E vs E only) (ablation) | Scoot metric · CUFS | MM1 resize instability: 0.200 -> 0.037 | Improved |
| add homogeneity feature to C+E (HEC vs CE) (ablation) | Scoot metric · CUFS | all meta-measures: 76.3 -> 76.3 | No clear effect |
| use a single feature instead of combining (ablation) | Scoot metric · CUFS | MM1 / MM3 | Mixed |
| replace C+E texture feature with GLRLM (ablation) | Scoot metric · CUFS | Jud: 76.3 -> 73.7 | Hurt |
| replace C+E texture feature with Gabor (ablation) | Scoot metric · CUFS | Jud: 76.3 -> 72.2 | Hurt |
| replace C+E with edge-based Canny or Sobel features (ablation) | Scoot metric · CUFS | Jud: 76.3 -> 27.8 | Hurt |
| use spatial grid k=1 instead of k=4 (ablation) | Scoot metric · CUFS | MM3 and MM4 | Mixed |
Closest papers by meaning
- SketchEvo: Leveraging Drawing Dynamics for Enhanced Image Synthesis (ICLR 2026 · similarity 0.86)
- Face Sketch Synthesis From Coarse to Fine (AAAI 2018 · similarity 0.86)
- One-Shot Reference-based Structure-Aware Image to Sketch Synthesis (AAAI 2025 · similarity 0.85)
- Sketch Down the FLOPs: Towards Efficient Networks for Human Sketch (CVPR 2025 · similarity 0.84)
- Evaluating Multiview Object Consistency in Humans and Image Models (NeurIPS 2024 · similarity 0.84)
- Visual Anagrams Reveal Hidden Differences in Holistic Shape Processing Across Vision Models (NeurIPS 2025 · similarity 0.84)
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.
