collaborators

5 papers

cs.LG2026

GeoEdit: Local Frames for Fast, Training-Free On-Manifold Editing in Diffusion Models

Yiming Zhang, Sitong Liu, Ke Li +3

Diffusion models are a leading paradigm for data generation, but training-free editing typically re-runs the full denoising trajectory for every edit strength, making iterative ref…

cs.CV2025

Improving Few-Shot Change Detection Visual Question Answering via Decision-Ambiguity-guided Reinforcement Fine-Tuning

Fuyu Dong, Ke Li, Di Wang +5

Change detection visual question answering (CDVQA) requires answering text queries by reasoning about semantic changes in bi-temporal remote sensing images. A straightforward appro…

cs.CV2025

A Dual-Branch Local-Global Framework for Cross-Resolution Land Cover Mapping

Peng Gao, Ke Li, Di Wang +4

Cross-resolution land cover mapping aims to produce high-resolution semantic predictions from coarse or low-resolution supervision, yet the severe resolution mismatch makes effecti…

cs.CV2025

RSVG-ZeroOV: Exploring a Training-Free Framework for Zero-Shot Open-Vocabulary Visual Grounding in Remote Sensing Images

Ke Li, Di Wang, Ting Wang +6

Remote sensing visual grounding (RSVG) aims to localize objects in remote sensing images based on free-form natural language expressions. Existing approaches are typically constrai…

cs.CV2025

Enhancing Cross-View Geo-Localization Generalization via Global-Local Consistency and Geometric Equivariance

Xiaowei Wang, Di Wang, Ke Li +6

Cross-view geo-localization (CVGL) aims to match images of the same location captured from drastically different viewpoints. Despite recent progress, existing methods still face tw…