isp-uv-es/dvl-instruct-eco-taco

DVL-Instruct (environmental assessment)

This is a repackaging, not a new dataset. It is DVL-Instruct (environmental assessment) by DynamicVL authors (Xuan et al.), USDA NAIP (imagery, US public domain), converted to TACO. Pixel values and labels are kept as released except where the description below says otherwise. All credit belongs to the original authors: if you use it, please cite them and follow their licence.

original dataset · paper · licence: apache-2.0

Repackaged into TACO by the Image and Signal Processing Group (ISP), Universitat de València, within the ELLIOT project.

Citation

Please cite the original work:

@inproceedings{xuan2025dynamicvl,
  title     = {DynamicVL: Benchmarking Multimodal Large Language Models for Dynamic City Understanding},
  author    = {Xuan, Weihao and Wang, Junjue and Qi, Heli and Chen, Zihang and Zheng, Zhuo and Zhong, Yanfei and Xia, Junshi and Yokoya, Naoto},
  booktitle = {Advances in Neural Information Processing Systems 38 (NeurIPS 2025) Datasets and Benchmarks Track},
  year      = {2025},
  eprint    = {2505.21076},
  archivePrefix = {arXiv}
}

About the data

11881 multiple-choice environmental-assessment questions over 8328 distinct NAIP stills at 1.0 m and 1024x1024 px: the eco_assessment half of DVL-Suite, in 5 question classes (co2, nighttime, o3, population, temperature).

8,328 samples · splits: test 1,550 · train 6,097 · validation 681 · tasks: visual-question-answering

Packaged as TACO v3.

Full description

Sample. One sample per frame, with the questions, options, answers and classes as index-aligned lists. 5,467 frames carry one question and the rest up to five, so one sample per question would store the same still up to five times, some 12 GB of duplicated pixels.

Scope. The sample axis here is a frame, while DVL-Instruct's change-series subset is organised by time series over the same pixels. data_id plus image_path is the join key to it, and a mixture drawing on both uses the same stills twice.

Bands. The near-infrared rides as its own optional leaf where NAIP acquired one; pre-2010 frames have none, a property of the programme rather than of this release.

Encoding. Re-encoded as COGs, losslessly and smaller than the source TIFFs (0.814x). No CRS and no transform; the stated 1.0 m GSD is on the bands.

Getting started

git clone --recursive https://github.com/OscarPellicer/taco
pip install -e "taco/python[ml]"     # builds the reader: C++23, CMake, Ninja, pkg-config, libcurl >= 7.83, OpenSSL >= 3

Read it straight from the Hub:

from huggingface_hub import snapshot_download
from taco.ml import Dataset, plot_sample

path = snapshot_download("isp-uv-es/dvl-instruct-eco-taco", repo_type="dataset")
ds = Dataset(path)                  # the repository is the container: parts + .tacocat
plot_sample(ds[0])

or from a local copy:

ds = Dataset("dvl-instruct-eco")
sample = ds[0]                      # {slot name: SlotValue}, arrays decoded
sample["rgb"].array.shape

Metadata without decoding anything:

import taco
taco.read("dvl-instruct-eco/.tacocat")              # one Arrow table, levels joined

Samples

What a sample contains

role	slot	holds	modality	detail
input	rgb	raster	optical	3 band(s), unit 1, requantised
input	nir	raster_series	optical	1 band(s), unit 1, requantised
input	questions	text_sequence	text
target	answers	text_sequence	text

Licence

apache-2.0

Providers: DynamicVL authors (Xuan et al.), USDA NAIP (imagery, US public domain)

Acknowledgements

TACO was designed by César Aybar and is specified at https://asterisk.coop/taco/spec/.

Built by Oscar Pellicer within the ELLIOT project at the Image and Signal Processing Group (ISP), Universitat de València.