Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 111 of the 111 references it cites were found in a registry or an index; 13 of the 46 sources its claims cite were checked for support.

YOLOv1 to v8: Unveiling Each Variant–A Comprehensive Review of YOLO

Open access
Hussain, Muhammad · 2024 · IEEE Access

Abstract

This paper implements a systematic methodological approach to review the evolution of YOLO variants. Each variant is dissected by examining its internal architectural composition, providing a thorough understanding of its structural components. Subsequently, the review highlights key architectural innovations introduced in each variant, shedding light on the incremental refinements. The review includes benchmarked performance metrics, offering a quantitative measure of each variant’s capabilities. The paper further presents the performance of YOLO variants across a diverse range of domains, manifesting their real-world impact. This structured approach ensures a comprehensive examination of YOLOs journey, methodically communicating its internal advancements and benchmarked performance before delving into domain applications. It is envisioned, the incorporation of concepts such as federated learning can introduce a collaborative training paradigm, where YOLO models benefit from training across multiple edge devices, enhancing privacy, adaptability, and generalisation.

Overview

CoChat Summary

## TL;DR

This paper is a literature review of YOLO from v1 to v8, focusing on architectural evolution, training strategies, benchmark performance, and applications across domains. It also argues that federated learning and edge deployment are promising directions for future YOLO development.

## Key findings

  • The review claims to be the first in-depth analysis of mainstream YOLO variants from YOLOv1 to YOLOv8 and says it spans more than 20 domains.
  • In the comparative table, reported performance ranges from 63.4% mAP / 45 FPS for YOLOv1 to 69.0% mAP / 52 FPS for YOLOv2, 57.9% mAP / 34 FPS for YOLOv3, 44.3% mAP / 65 FPS for YOLOv4, 50.7% mAP / 200 FPS for YOLOv5, 52.5% mAP / 29 FPS for YOLOv6, 56.8% mAP / 5–160 FPS for YOLOv7, and 53.9% mAP / 280 FPS for YOLOv8.
  • YOLOv6 is reported to have eight scaled variants, with the largest variant reaching 57.2% AP on MS COCO test-dev 2017 at about 29 FPS on an NVIDIA Tesla T4.
  • Example application results in the survey include 99.7% Accuracy for breast mass detection, 96.42% F1 score for suspicious activity detection, 95.1% MAP for wind turbine blade defect detection, and 92% F1-score for petrochemical pipeline defect detection.

## Methods

  • The paper is a systematic review of YOLO variants from YOLOv1 through YOLOv8, organized around architecture, training strategies, performance benchmarks, and real-world applications.
  • It draws on benchmark datasets and comparative tables mentioned in the text, especially Pascal VOC and MS COCO; it also discusses dataset diversity and augmentation methods such as Mosaic, MixUp, random affine, and others.
  • Main analysis consists of qualitative architectural comparison plus quantitative summary tables of mAP/FPS and application-specific results across domains.

## Limitations

  • The paper states that YOLO still faces challenges in occlusions and clutter, scale variations, fine-grained object detection, domain adaptation, biases, and real-time edge deployment.
  • The review itself does not report a formal experimental protocol beyond summarizing published results; this is evident from the text because it is a survey and not an original controlled study.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1109/access.2024.3378568

Cites

111 of 111 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

11 Valid · 20 Not valid · 27 Not checked · 80 Own results and interpretations

27 more claims

Create a free account to see all 138 claims and to verify the 9 references still open.

See all claims — free account
  • It may state two separate claims as one. 56
  • Probably not a substantial claim. 73
  • The quote only partly states this claim. 2

Create a free account to see these claims.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

2 of 2

Citations

Cites

Show all 105 references

Research works printed but not linked: 1

  1. Wang, 2022, YOLOv6

Cited by

See all 492

Metadata

Authors
Hussain, M.
Affiliations
University of Huddersfield (GB)
Corresponding author
Hussain, Muhammad
Venue
IEEE Access
Published
2024
Volume
12
Pages
42816-42833
ISSN
2169-3536 (ISSN-L)
OpenAlex
W4392939809
Cited by
492
References
105
Citations by year
2026: 168 · 2025: 253 · 2024: 71
FWCI
68.53
Type
journal article
Language
en
Open access
Locations
  • IEEE Access (publishedVersion, cc-by-nc-nd)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Huddersfield Research Portal (University of Huddersfield) (submittedVersion, cc-by-nc-nd)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
  • Life Sciences › Agricultural and Biological Sciences › Plant Science › Smart Agriculture and AI
Keywords
YOLO, object detection, YOLOv8, benchmark performance, federated learning, edge devices, privacy-preserving training

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: YOLOv1 to v8: Unveiling Each Variant–A Comprehensive Review of YOLO. https://app.cochat.ai/papers/10.1109/access.2024.3378568
The paper
Hussain, M. (2024). YOLOv1 to v8: Unveiling Each Variant–A Comprehensive Review of YOLO. IEEE Access. https://doi.org/10.1109/access.2024.3378568

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.