ICCV 2025 - 3 Papers Accepted
- halosaber
- Jun 25
- 1 min read
We have three papers accepted for ICCV 2025 to be held from Oct 19 to 23th, 2025, Honolulu, Hawaii, USA.
Title: Bridging the Skeleton-Text Modality Gap: Diffusion-Powered Modality Alignment for Zero-shot Skeleton-based Action Recognition
Authors: Jeonghyeok Do, Munchurl Kim† (†: Corresponding authors)
Project Page:
Title: PAN-Crafter: Learning Modality-Consistent Alignment for PAN-Sharpening
Authors: Jeonghyeok Do, Sungpyo Kim, Geunhyuk Youk, Jaehyup Lee† and Munchurl Kim† (†: Co-corresponding authors)
Project Page:
Title: One Look is Enough: Seamless Patchwise Refinement for Zero-Shot Monocular Depth Estimation
Authors: Byeongjun Kwon, Munchurl Kim† (†: Corresponding authors)
Project Page:
Comments