Abstract:With the continuous advancement of remote sensing observation technology, wide-area observation, and high-resolution imaging make remote sensing images contain a large number of dense tiny objects. The detection of dense tiny objects is a very challenging task since these objects are with very low resolution and might stick together. Existing work lacks further exploration of the contextual scene information and inherent characteristics of dense tiny objects, which are crucial for performance improvement of dense tiny object detection (DTOD). In this work, we propose a novel scene contextualized detection network (SCDNet) by decoupling scene contextual information through a dedicated scene classification subnetwork (SCN), thereby enabling an enhanced exploration of the relationship between tiny objects and their surrounding environments. In particular, we design a lightweight scene context guided fusion module (SCGM) in SCDNet to incorporate scene context information around dense tiny objects more effectively. Moreover, we further develop the scene context guided foreground enhancement module (FEM) to suppress the background information while enhancing the foreground information based on the scene information. In addition, this research field still lacks a large-scale benchmark dataset with dense tiny objects, which is crucial for the training and comprehensive evaluation of detection methods. To this end, we construct a large-scale dataset for DTOD. It contains 11600 images with 1019800 instances, the average absolute size of objects is smaller than 13 pixels, and each image contains 88 objects on average. Extensive experiments are conducted on the proposed dataset, and the results demonstrate the superiority and effectiveness of SCDNet compared to existing methods. The dataset and evaluation code are available at https://github.com/mmic-lcl.

Dense Tiny Object Detection: A Scene Context Guided Approach and a Unified Benchmark

Tiny object detection with context enhancement and feature purification

(Sod)-O-3: Single Stage Small Object Detector From Scratch For Remote Sensing Images

A Tiny Object Detection Method Based on Explicit Semantic Guidance for Remote Sensing Images

Towards Large-Scale Small Object Detection: Survey and Benchmarks.

Tiny Object Tracking: A Large-scale Dataset and A Baseline

Oriented Tiny Object Detection: A Dataset, Benchmark, and Dynamic Unbiased Learning

Towards densely clustered tiny pest detection in the wild environment

Multistage Enhancement Network for Tiny Object Detection in Remote Sensing Images

Tiny-DSOD: Lightweight Object Detection for Resource-Restricted Usages

Multi-stage Enhancement Network for Tiny Object Detection in Remote Sensing Images

Small object detection leveraging density‐aware scale adaptation

Feature Rescaling and Fusion for Tiny Object Detection

STODNet: Sparse Convolution for Super Tiny Object Detection from Remote Sensing Image

Visible and Clear: Finding Tiny Objects in Difference Map

SWIN-TOD: Smooth Wasserstein Distance and Instance-level Neighboring Enhancement for Remote Sensing Tiny Object Detection

Beyond Context: Exploring Semantic Similarity for Small Object Detection in Crowded Scenes

A Survey of the Four Pillars for Small Object Detection: Multiscale Representation, Contextual Information, Super-Resolution, and Region Proposal

CSPGNet: Cross-Scale Spatial Perception Guided Network for Tiny Object Detection in Remote Sensing Images

Small and Dense Commodity Object Detection with Multi-Scale Receptive Field Attention

Regression-Guided Refocusing Learning With Feature Alignment for Remote Sensing Tiny Object Detection