Pith. sign in

REVIEW 9 cited by

A Normalized Gaussian Wasserstein Distance for Tiny 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 2110.13389 v2 pith:GXWZP6GH submitted 2021-10-26 cs.CV

Jinwang Wang , Chang Xu , Wen Yang , Lei Yu This is my paper
classification cs.CV
keywords tinymetricobjectdetectiondistancegaussianobjectswasserstein
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Detecting tiny objects is a very challenging problem since a tiny object only contains a few pixels in size. We demonstrate that state-of-the-art detectors do not produce satisfactory results on tiny objects due to the lack of appearance information. Our key observation is that Intersection over Union (IoU) based metrics such as IoU itself and its extensions are very sensitive to the location deviation of the tiny objects, and drastically deteriorate the detection performance when used in anchor-based detectors. To alleviate this, we propose a new evaluation metric using Wasserstein distance for tiny object detection. Specifically, we first model the bounding boxes as 2D Gaussian distributions and then propose a new metric dubbed Normalized Wasserstein Distance (NWD) to compute the similarity between them by their corresponding Gaussian distributions. The proposed NWD metric can be easily embedded into the assignment, non-maximum suppression, and loss function of any anchor-based detector to replace the commonly used IoU metric. We evaluate our metric on a new dataset for tiny object detection (AI-TOD) in which the average object size is much smaller than existing object detection datasets. Extensive experiments show that, when equipped with NWD metric, our approach yields performance that is 6.7 AP points higher than a standard fine-tuning baseline, and 6.0 AP points higher than state-of-the-art competitors. Codes are available at: https://github.com/jwwangchn/NWD.

Discussion (0). Sign in to comment.

Forward citations

Cited by 9 Pith papers

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

  1. TimeLens2: Generalist Video Temporal Grounding with Multimodal LLMs

    cs.CV 2026-07 conditional novelty 7.0 of 10

    TimeLens2 shows that a compact video MLLM can localize multiple evidence intervals in long videos by training on verified interval labels and a Wasserstein-based time-distance reward.

  2. TargetFinder: Detecting Widgets from Pixels on Desktop Interfaces

    cs.HC 2026-07 conditional novelty 6.0 of 10

    A fine-tuned YOLO pipeline on a new 520-screenshot, 38,000-widget dataset detects desktop GUI widgets from pixels and drives system-wide Bubble Cursor and Semantic Pointing.

  3. MSG-Loc: Multi-Label Likelihood-based Semantic Graph Matching for Object-Level Global Localization

    cs.RO 2025-12 conditional novelty 6.0 of 10

    Object-level global localization becomes more robust to semantic ambiguity by matching multi-label confidence distributions and propagating neighbor likelihoods across semantic graphs.

  4. An Uncertainty-aware DETR Enhancement Framework for Object Detection

    cs.CV 2025-07 conditional novelty 6.0 of 10

    Adding a Gaussian-box Gromov-Wasserstein loss and Bayes-risk-based refinement to DETR detectors improves their AP on COCO and leukocyte datasets while producing localization uncertainty estimates.

  5. 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...

  6. ICME 2026 Grand Challenge on Cross-Scenario Defect Detection and Fine-Grained Severity Grading for High-Precision Manufacturing

    cs.CV 2026-07 accept novelty 5.0 of 10

    A new industrial wafer-defect challenge and dataset for cross-scenario instance detection and ordinal severity grading, with leaderboards from 21 finalist teams.

  7. Inter-Class Relational Loss for Small Object Detection: A Case Study on License Plates

    cs.CV 2025-08 unverdicted novelty 5.0 of 10

    A new relational loss adds a penalty when a plate's predicted box misses its car, reportedly boosting mAP on two detectors.

  8. CEM-FBGTinyDet: Context-Enhanced Foreground Balance with Gradient Tuning for tiny Objects

    cs.CV 2025-06 reject novelty 4.0 of 10

    A tiny-object detector with high-to-low level feature fusion and a sigmoid-weighted L1/L2 loss reports +1.3 AP on AI-TOD, but the loss gradient claims are contradicted by the paper's own equations.

  9. Open-Vocabulary Object Detection in UAV Imagery: A Review and Future Perspectives

    cs.CV 2025-07 conditional novelty 3.0 of 10

    A structured review that divides aerial open-vocabulary detection methods into pseudo-labeling and CLIP-driven integration families and catalogs the missing benchmarks in the field.

Pith tools