ESOD: Efficient Small Object Detection on High-Resolution Images

Enlarging input images is a straightforward and effective approach to promote small object detection. However, simple image enlargement is significantly expensive on both computations and GPU memory. In fact, small objects are usually sparsely distributed and locally clustered. Therefore, massive fe...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on image processing 2025, Vol.34, p.183-195
Main Authors: Liu, Kai, Fu, Zhihang, Jin, Sheng, Chen, Ze, Zhou, Fan, Jiang, Rongxin, Chen, Yaowu, Ye, Jieping
Format: Article
Language:English
Subjects:
Computation
Costs
Detectors
Distributed memory
Effectiveness
Feature extraction
Feeds
filter-then-detect
Graphics processing units
Head
High resolution
high-resolution detection
Image resolution
Neck
Object detection
Object recognition
Small object detection
sparse detection
Target detection
Visualization
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Enlarging input images is a straightforward and effective approach to promote small object detection. However, simple image enlargement is significantly expensive on both computations and GPU memory. In fact, small objects are usually sparsely distributed and locally clustered. Therefore, massive feature extraction computations are wasted on the non-target background area of images. Recent works have tried to pick out target-containing regions using an extra network and perform conventional object detection, but the newly introduced computation limits their final performance. In this paper, we propose to reuse the detector's backbone to conduct feature-level object-seeking and patch-slicing, which can avoid redundant feature extraction and reduce the computation cost. Incorporating with a sparse detection head, we are able to detect small objects on high-resolution inputs (e.g., 1080P or larger) for superior performance. The resulting Efficient Small Object Detection (ESOD) approach is a generic framework, which can be applied to both CNN- and ViT-based detectors to save the computation and GPU memory costs. Extensive experiments demonstrate the efficacy and efficiency of our method. In particular, our method consistently surpasses the SOTA detectors by a large margin (e.g., 8% gains on AP) on the representative VisDrone, UAVDT, and TinyPerson datasets. Code will be made public soon.
ISSN:1057-7149
1941-0042
DOI:10.1109/TIP.2024.3501853