SARDet-100K

Dataset Information
Modalities
Graphs
Languages
English, Chinese
Introduced
2024
License
Attribution-NonCommercial 4.0 International
Homepage
Official Website
Contents

Overview

The SARDet-100K dataset encompasses a total of 116,598 images, and 245,653 instances distributed across six categories: Aircraft, Ship, Car, Bridge, Tank, and Harbor. SARDet100K dataset stands as the first large-scale SAR object detection dataset, comparable in size to the widely used COCO dataset (118K images). The scale and diversity of the SARDet-100K dataset provide researchers with robust training and evaluation for advancing SAR object detection algorithms and techniques, fostering the development of SOTA models in this domain.

Variants: SARDet-100K

Associated Benchmarks

This dataset is used in 1 benchmark:

Recent Benchmark Submissions

Task Model Paper Date
2D Object Detection DenoDet DenoDet: Attention as Deformable Multi-Subspace … 2024-06-05
2D Object Detection SARATR-X SARATR-X: Toward Building A Foundation … 2024-05-15
2D Object Detection MSFA (GFL+R50) SARDet-100K: Towards Open-Source Benchmark and … 2024-03-11
2D Object Detection MSFA (Deformable DETR) SARDet-100K: Towards Open-Source Benchmark and … 2024-03-11
2D Object Detection MSFA (F-RCNN+R50) SARDet-100K: Towards Open-Source Benchmark and … 2024-03-11
2D Object Detection MSFA (F-RCNN+ConvNext-T) SARDet-100K: Towards Open-Source Benchmark and … 2024-03-11
2D Object Detection Sparse R-CNN Sparse R-CNN: End-to-End Object Detection … 2020-11-25
2D Object Detection Deformable DETR Deformable DETR: Deformable Transformers for … 2020-10-08
2D Object Detection FCOS FCOS: Fully Convolutional One-Stage Object … 2019-04-02
2D Object Detection Grid RCNN Grid R-CNN 2018-11-29
2D Object Detection Cascade R-CNN Cascade R-CNN: Delving into High … 2017-12-03
2D Object Detection RetinaNet Focal Loss for Dense Object … 2017-08-07
2D Object Detection F-RCNN Faster R-CNN: Towards Real-Time Object … 2015-06-04

Research Papers

Recent papers with results on this dataset:

External Links: