Illustration of feature matching of UFM. When working with specific data, the pre-trained backbone is frozen, and only the corresponding modal assistants need to be fine-tuned for feature matching. The multi-modal assistants contain both same-modal assistants and different-modal matching assistants.

Illustration of feature matching of UFM. When working with specific data, the pre-trained backbone is frozen, and only the corresponding modal assistants need to be fine-tuned for feature matching. The multi-modal assistants contain both same-modal assistants and different-modal matching assistants.

<p>Illustration of feature matching of UFM. When working with specific data, the pre-trained backbone is frozen, and only the corresponding modal assistants need to be fine-tuned for feature matching. The multi-modal assistants contain both same-modal assistants and different-modal matching as...

Full description

Saved in:
Bibliographic Details
Main Author: Yide Di (20969124) (author)
Other Authors: Yun Liao (160524) (author), Hao Zhou (136535) (author), Kaijun Zhu (18283913) (author), Qing Duan (541846) (author), Junhui Liu (2063140) (author), Mingyu Lu (2333083) (author)
Published: 2025
Subjects:
Sociology
Science Policy
Space Science
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
multimodal image applications
image feature matching
experimental results demonstrate
data augmentation algorithm
across different modals
ufm exhibits versatility
div >< p
imbalanced modal datasets
specific modals
specific datasets
sparse data
wide spectrum
ufm excels
ufm </
trained model
staged pre
remains challenging
modal images
foundational task
computer vision
Tags: Add Tag
No Tags, Be the first to tag this record!

Similar Items