Meitu MT Lab Presents CFT for Stable Portrait Relighting at ECCV 2026
jiqizhixin · x · 2026-08-27
Meitu Imaging Technology Lab (MT Lab) presented CFT (Consistent Feature Transport), accepted at ECCV 2026. This method reframes portrait relighting as a lighting-consistent feature transport problem. Built on Rectified Flow, CFT jointly models noise, source image, and target image distributions, explicitly learning lighting-specific feature displacement rather than generic image-to-image translation. This approach preserves identity and structure stability while altering only the illumination. CFT achieves state-of-the-art scores across all four relighting metrics. The concept of consistent feature transport also generalizes to style transfer, improving content structure and perceptual fidelity.
More from Multimodal
- Fal model generates long clips in 15 seconds, a transformative speed — JenniferHli · 2026-08-27
- FIRM-Video: Reliable Reward Models via Checklist Verification — VisionXLab · 2026-08-27
- Surflo: Generate Consistent 3D Surfaces from Arbitrary Photos — jonstephens85 · 2026-08-27
- NVIDIA Releases ARDY: Real-Time Interactive Human Motion Generation Model — rsasaki0109 · 2026-08-27
- 15-Second 768p Video Generated in 5.12 Seconds — umesh_ai · 2026-08-27
- Visual General Intelligence White Paper: Vision as a Pathway to AGI — zhenjun_zhao · 2026-08-27