Pith. sign in

REVIEW 4 cited by

Augmentation for small object detection

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1902.07296 v1 pith:AGITLBAY submitted 2019-02-19 cs.CV

Mate Kisantal , Zbigniew Wojna , Jakub Murawski , Jacek Naruniec , Kyunghyun Cho This is my paper
classification cs.CV
keywords objectssmalldetectionimagesobjectaugmentationcococontaining
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

In recent years, object detection has experienced impressive progress. Despite these improvements, there is still a significant gap in the performance between the detection of small and large objects. We analyze the current state-of-the-art model, Mask-RCNN, on a challenging dataset, MS COCO. We show that the overlap between small ground-truth objects and the predicted anchors is much lower than the expected IoU threshold. We conjecture this is due to two factors; (1) only a few images are containing small objects, and (2) small objects do not appear enough even within each image containing them. We thus propose to oversample those images with small objects and augment each of those images by copy-pasting small objects many times. It allows us to trade off the quality of the detector on large objects with that on small objects. We evaluate different pasting augmentation strategies, and ultimately, we achieve 9.7\% relative improvement on the instance segmentation and 7.1\% on the object detection of small objects, compared to the current state of the art method on MS COCO.

Discussion (0). Sign in to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Adaptive Spectrum-Aware Feature Disentangled Network for Small Object Detection

    cs.CV 2026-06 unverdicted novelty 6.0 of 10

    SFDNet improves small object detection by decomposing features into spectral components to remove distractors and distilling class prototypes for semantic consistency.

  2. Physics-Informed Super-Resolution of Atmospheric Data

    cs.LG 2026-07 reject novelty 5.0 of 10

    Adding multi-scale hydrostatic-primitive-equation losses to atmospheric super-resolution models improves reported physical-consistency scores and some reconstruction/event-detection metrics, but the metric and constra...

  3. A Data-Driven RetinaNet Model for Small Object Detection in Aerial Images

    cs.CV 2025-09 conditional novelty 5.0 of 10

    A data-driven RetinaNet variant that selects feature pyramid levels and anchor sizes from training data improves small-object detection in aerial bird imagery.

  4. DyFrDet: Towards Accurate Small Object Detection via Dynamic Frequency Suppression with Label Disambiguation

    cs.CV 2026-08 conditional novelty 4.0 of 10

    DyFrDet reports state-of-the-art small-object detection on AI-TOD and SODA by dynamically masking frequency bands and down-weighting ambiguous regression labels.

Pith tools