Datasets › DOTA

DOTA (Dataset for Object deTection in Aerial Images)

Introduced by Gui-Song Xia et al. in DOTA: A Large-scale Dataset for Object Detection in Aerial Images1 Jan 2018 archive 2025-07-28

DOTA is a large-scale dataset for object detection in aerial images. It can be used to develop and evaluate object detectors in aerial images. The images are collected from different sensors and platforms. Each image is of the size in the range from 800 × 800 to 20,000 × 20,000 pixels and contains objects exhibiting a wide variety of scales, orientations, and shapes. The instances in DOTA images are annotated by experts in aerial image interpretation by arbitrary (8 d.o.f.) quadrilateral. We will continue to update DOTA, to grow in size and scope to reflect evolving real-world conditions. Now it has three versions:

DOTA-v1.0 contains 15 common categories, 2,806 images and 188, 282 instances. The proportions of the training set, validation set, and testing set in DOTA-v1.0 are 1/2, 1/6, and 1/3, respectively.

DOTA-v1.5 uses the same images as DOTA-v1.0, but the extremely small instances (less than 10 pixels) are also annotated. Moreover, a new category, ”container crane” is added. It contains 403,318 instances in total. The number of images and dataset splits are the same as DOTA-v1.0. This version was released for the DOAI Challenge 2019 on Object Detection in Aerial Images in conjunction with IEEE CVPR 2019.

DOTA-v2.0 collects more Google Earth, GF-2 Satellite, and aerial images. There are 18 common categories, 11,268 images and 1,793,658 instances in DOTA-v2.0. Compared to DOTA-v1.5, it further adds the new categories of ”airport” and ”helipad”. The 11,268 images of DOTA are split into training, validation, test-dev, and test-challenge sets. To avoid the problem of overfitting, the proportion of training and validation set is smaller than the test set. Furthermore, we have two test sets, namely test-dev and test-challenge. Training contains 1,830 images and 268,627 instances. Validation contains 593 images and 81,048 instances. We released the images and ground truths for training and validation sets. Test-dev contains 2,792 images and 353,346 instances. We released the images but not the ground truths. Test-challenge contains 6,053 images and 1,090,637 instances.

Source: https://captain-whu.github.io/DOTA/index.html Image Source: https://captain-whu.github.io/DOTA/

Benchmarks archive 2025-07-28

All 2 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

Papers archive 2025-07-28

30 shown of 47 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 293. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
LEGNet: Lightweight Edge-Gaussian Driven Network for Low-Quality Remote Sensing Image Object Detection 1 1 18 Mar 2025 not harvested
LWGANet: A Lightweight Group Attention Backbone for Remote Sensing Visual Tasks 1 1 17 Jan 2025 not harvested
Strip R-CNN: Large Strip Convolution for Remote Sensing Object Detection 3 2 7 Jan 2025 not harvested
DecoupleNet: A Lightweight Backbone Network With Efficient Feature Decoupling for Remote Sensing Visual Tasks 1 1 23 Sep 2024 not harvested
Projecting Points to Axes: Oriented Object Detection via Point-Axis Representation 1 1 11 Jul 2024 not harvested
Category-Aware Dynamic Label Assignment with High-Quality Oriented Proposal 0 1 3 Jul 2024 not harvested
MTP: Advancing Remote Sensing Foundation Model via Multi-Task Pretraining 2 3 20 Mar 2024 ran 4 of 5 samples (1 unverified)
LSKNet: A Foundation Lightweight Backbone for Remote Sensing 2 3 18 Mar 2024 not harvested
MoCaE: Mixture of Calibrated Experts Significantly Improves Object Detection 1 1 26 Sep 2023 ran 1 of 1 samples (0 unverified; 1 pointer-only for licence)
Spatial Transform Decoupling for Oriented Object Detection 1 1 21 Aug 2023 not harvested
Adaptive Rotated Convolution for Rotated Object Detection 1 1 14 Mar 2023 ran 3 of 3 samples (0 unverified)
RTMDet: An Empirical Study of Designing Real-Time Object Detectors 14 2 14 Dec 2022 ran 3 of 20 samples (17 unverified)
PP-YOLOE-R: An Efficient Anchor-Free Rotated Object Detector 2 4 4 Nov 2022 not harvested
Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model 2 2 8 Aug 2022 ran 3 of 4 samples (1 unverified)
An Empirical Study of Remote Sensing Pretraining 2 4 6 Apr 2022 not harvested
Few Could Be Better Than All: Feature Sampling and Grouping for Scene Text Detection 0 1 29 Mar 2022 not harvested
EAutoDet: Efficient Architecture Search for Object Detection 0 1 21 Mar 2022 ran 0 of 2 samples (2 unverified; 2 pointer-only for licence)
Anchor Retouching via Model Interaction for Robust Object Detection in Aerial Images 1 1 13 Dec 2021 not harvested
A General Gaussian Heatmap Label Assignment for Arbitrary-Oriented Object Detection 1 1 27 Sep 2021 not harvested
Oriented R-CNN for Object Detection 5 1 12 Aug 2021 not harvested
Beyond Bounding-Box: Convex-Hull Feature Adaptation for Oriented and Densely Packed Object Detection 2 1 19 Jun 2021 not harvested
Learning High-Precision Bounding Box for Rotated Object Detection via Kullback-Leibler Divergence 2 1 3 Jun 2021 not harvested
Oriented RepPoints for Aerial Object Detection 2 1 24 May 2021 ran 0 of 2 samples (2 unverified)
TricubeNet: 2D Kernel-Based Object Representation for Weakly-Occluded Oriented Object Detection 1 1 23 Apr 2021 not harvested
Optimization for Arbitrary-Oriented Object Detection via Representation Invariance Loss 1 1 22 Mar 2021 not harvested
ReDet: A Rotation-equivariant Detector for Aerial Object Detection 4 1 13 Mar 2021 not harvested
Rethinking Rotated Object Detection with Gaussian Wasserstein Distance Loss 2 1 28 Jan 2021 not harvested
CFC-Net: A Critical Feature Capturing Network for Arbitrary-Oriented Object Detection in Remote Sensing Images 1 1 18 Jan 2021 not harvested
Dynamic Anchor Learning for Arbitrary-Oriented Object Detection 2 1 8 Dec 2020 not harvested
Dense Label Encoding for Boundary Discontinuity Free Rotation Detection 3 1 19 Nov 2020 ran 1 of 1 samples (0 unverified; 1 pointer-only for licence)

The full list of 47 is in the JSON twin.

Dataset loaders archive 2025-07-28

No loader listed in the archive.

Tasks archive 2025-07-28

License archive 2025-07-28

Custom (non-commercial)

Modalities archive 2025-07-28

Languages archive 2025-07-28

No language tagged.

Variants archive 2025-07-28

  • DOTA
  • DOTA 1.5
  • DOTA 1.0
  • DOTA 2.0

4 variant names, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections