ReFlowSET: representation-aligned latent flow matching for SAR-to-EO translation
KAIST · hf · 2026-09-02
KAIST's ReFlowSET selects a latent codec via joint SAR-EO reconstruction, then trains a small conditional DiT with dual-stream conditioning and frozen vision-model alignment for high-fidelity SAR-to-EO image translation.
More from Multimodal
- You can distill consistent surface meshes from the Atlas world model — MatthewChang · 2026-09-02
- New paper on Native Unified Multimodal Models — liuziwei7 · 2026-09-02
- World Labs' Atlas could speed up humanoid training via real2sim2real — CyberRobooo · 2026-09-02
- Opus 5 demonstrates breathing exercises and voice synthesis control — repligate · 2026-09-02
- Fable 5 vs 5.1 self-portrait comparison re-run without the new system prompt — repligate · 2026-09-02
- Field Life demonstrates 3D generation capabilities — BertChakovsky · 2026-09-02