Search

Results for "Image Processing"

AI helps create street maps from satellite imagery - GPS World

QCRI · Mar 5

Artificial intelligence is increasingly being utilized to generate and update detailed street maps directly from satellite imagery. This technology automates the identification and extraction of roads, buildings, and other geographical features, significantly reducing the manual effort traditionally required for cartography. It offers a solution for rapidly updating maps in dynamic urban environments and for mapping remote or previously uncharted areas. Why it matters: This advancement improves the accuracy and timeliness of geospatial data, essential for navigation, urban planning, disaster management, and infrastructure development globally.

The Prism Hypothesis: Harmonizing Semantic and Pixel Representations via Unified Autoencoding

arXiv · Dec 22

The paper introduces the Prism Hypothesis, which posits a correspondence between an encoder's feature spectrum and its functional role, with semantic encoders capturing low-frequency components and pixel encoders retaining high-frequency information. Based on this, the authors propose Unified Autoencoding (UAE), a model that harmonizes semantic structure and pixel details using a frequency-band modulator. Experiments on ImageNet and MS-COCO demonstrate that UAE effectively unifies semantic abstraction and pixel-level fidelity, achieving state-of-the-art performance.