Research map / Image generation / Editing and style transfer
TraceTrans: Translation and Spatial Tracing for Surgical Prediction
Xiyu Luo, Haodong Li, Xinxing Cheng, He Zhao, Yang Hu, Xuan Song, Tianyang 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 |
|---|---|---|---|
| add two-stream architecture (no constraint) (ablation) | single-stream baseline · Edge correspondence | SSIM (%): 55.58 -> 58.50 | Improved |
| add cross-domain constraint to two-stream (ablation) | TraceTrans · Edge correspondence | SSIM (%): 58.50 -> 59.48 | Improved |
| add deformation stream (ablation) | TraceTrans · Face Cosmetology | MAE: 12.88 -> 12.03 | Improved |
| add cross-domain constraint (ablation) | TraceTrans · Face Cosmetology | MAE: 12.03 -> 10.64 | Improved |
| lower trans proportion alpha from 1 to 0.5 (ablation) | TraceTrans · Face Cosmetology | Masked Dice (%): 44.72 -> 51.10 | Improved |
| lower trans proportion alpha from 1 to 0.5 (ablation) | TraceTrans · Brain MRI | SSIM (%): 74.01 -> 75.97 | Improved |
| lower alpha from 0.5 to 0.25 (ablation) | TraceTrans · Brain MRI | SSIM (%): 75.97 -> 75.57 | Hurt |
| increase cross-domain strength gamma from 0 to 1.0 (ablation) | TraceTrans · Face Cosmetology | Masked SSIM (%): 83.00 -> 85.16 | Improved |
| increase cross-domain strength gamma from 0 to 1.0 (ablation) | TraceTrans · Face Cosmetology | Masked Dice (%): 45.60 -> 51.10 | Improved |
| TraceTrans vs best baseline Pix2Pix | Pix2Pix · Face Cosmetology | SSIM (%): 79.93 -> 82.92 | Improved |
| TraceTrans vs MUNIT on PSNR | MUNIT · Brain MRI | PSNR: 31.52 -> 31.48 | No clear effect |
Closest papers by meaning
- How Well Do Supervised 3D Models Transfer to Medical Imaging Tasks? (ICLR 2024 · similarity 0.84)
- Knowledge Transfer Scaling Laws for 3D Medical Imaging (NeurIPS 2026 · similarity 0.83)
- Trade-Offs in Fine-Tuned Diffusion Models between Accuracy and Interpretability (AAAI 2024 · similarity 0.83)
- LiFT: Lifted Inter-slice Feature Trajectories for 3D Image Generation from 2D Generators (NeurIPS 2026 · similarity 0.83)
- Bidirectional Recurrence for Cardiac Motion Tracking with Gaussian Process Latent Coding (NeurIPS 2024 · similarity 0.82)
- Better Tokens for Better 3D: Advancing Vision-Language Modeling in 3D Medical Imaging (NeurIPS 2025 · similarity 0.82)
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.
