#Cross-Modal Visual Perception Consistency: A Language-Enhanced Approach for Heterogeneous Change Detection

We construct LEVPC, a heterogeneous change detection method that incorporates language information to enhance visual feature learning.Leveraging language as a bridge, it facilitates cross-modal relationship discovery and anchors feature consistency transformation, improving change identification under the condition of heterogeneous scenarios.

The code will be made public soon.