Dataset Hub

The LEVIR‑CC dataset is a large dataset for remote‑sensing image change caption generation, specifically supporting research on change captioning using dual‑branch Transformers.

We provide a database of hyperspectral images (HSIs) supporting our research. The images cover a variety of real‑world materials and objects. The database is publicly released to the research community. Detailed information can be found in our manuscript.

This is a multimodal remote‑sensing dataset for Hunan Province in 2017, comprising Sentinel‑2, Sentinel‑1, and SRTM data. It contains 400 training images (256×256), and 50 validation and test images. The TRI in the training set is computed from SRTM via GDAL and can be used for knowledge reconstruction.

LHRS‑Align is a large‑scale, semantically rich and feature‑diverse remote‑sensing image‑text alignment dataset. It leverages volunteer geographic information (VGI) from OpenStreetMap and remote‑sensing images from Google Earth, containing 1.15 million high‑quality RS image‑text pairs.

This dataset was collected by JPL's airborne Visible/Infrared Imaging Spectrometer (AVIRIS) on August 20, 1992, at an altitude of 20,000 m over Moffett Field, California. It contains an image cube that illustrates the instrument's data output. The top of the cube is a false‑color image highlighting the structure of water and evaporation ponds, while the side displays the edges of all 224 AVIRIS spectral channels. The dataset spans the visible‑to‑infrared spectral range and shows strong responses in specific areas due to the presence of red brine shrimp in the evaporation ponds.