Papers › YOLOv3: An Incremental Improvement
YOLOv3: An Incremental Improvement
Joseph Redmon, Ali Farhadi
We present some updates to YOLO! We made a bunch of little design changes to make it better. We also trained this new network that's pretty swell. It's a little bigger than last time but more accurate. It's still fast though, don't worry. At 320x320 YOLOv3 runs in 22 ms at 28.2 mAP, as accurate as SSD but three times faster. When we look at the old .5 IOU mAP detection metric YOLOv3 is quite good. It achieves 57.9 mAP@50 in 51 ms on a Titan X, compared to 57.5 mAP@50 in 198 ms by RetinaNet, similar performance but 3.8x faster. As always, all the code is online at https://pjreddie.com/yolo/
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="1804.02767")
Code
Syntology Ran 18 of 124 code samples harvested from 22 repositories linked to this paper; 106 have no recorded run. Of those that ran: 4 ran · honoured contract; 2 ran · violated contract; 7 ran · our draft was wrong; 4 ran · fixture could not drive it; 1 ran with no contract checked.
By repository: community (archive-listed): 114 samples from 22 repositories, 10 ran; 10 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.
311 repositories listed; official and paper-mentioned ones first.
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
124 samples harvested; 18 ran; 4 honoured the contract we drafted; 106 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.
Licence: 19 of the 124 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.
Harvested from 22 repositories linked to this paper, official or community; each sample names its own and says which. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.
Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.
Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.
988a3c854b1b13d0 · report
ce470c95b08cd7db · report
6044c07c7c65ad55 · report
df23da82fbd7ff7b · report
ed702a199df9d0db · report
ec338718ed4c4fe1 · report
fa99a26b6ba8f205 · report
b9e24c3e3e0e88a7 · report
5f33cd0434b7657c · report
3b40b5666c482d42 · report
794b3117107512ef · report
6e9f112d5b789a72 · report
bf73538d10204e2d · report
7a50e764fb28b748 · report
521c558b71256b1d · report
73c23db542d9513e · report
2e3a1b43bdfce447 · report
fe9bc9c30086c04f · report
a28227d670ba6b7e · report
5bac87cf119ddef3 · report
e460868d0e106ef7 · report
91b4702f6af52c0c · report
ef2908ceb8d3dbd6 · report
acdd9e1ef705fe40 · report
8749f6c6e44471e8 · report
d6ae8e6a9c267f7b · report
613f27d12519398d · report
74010f265a9ac192 · report
67f851a5d61759bc · report
9c414d9463c241cd · report
8c6d05a8b92e5e5c · report
723dd698cf920673 · report
bd2be659aa2aba95 · report
e3f1588842c9401c · report
f9299ec3aa73a4dd · report
8e9eb6523c73aa6e · report
fe8d09aa10388d81 · report
0baeb54a2e10e06c · report
d1cfcce6c4def27a · report
9d9a2b61c5bcbecb · report
dbdb9af98dfff0c1 · report
2ae73fdbbb2cb8ba · report
4958de84124bd877 · report
7841edd8ae71f548 · report
ec014668804d4c46 · report
eb22d53f46cdff31 · report
d9196437ddef4b60 · report
ae5f107e8e1efb7d · report
688442602fea3347 · report
a3c4fa7fd984d56b · report
f0656af9cfc0e4cc · report
b13acc0922dade5c · report
410bf279e17ed843 · report
f938002f1fb9665f · report
2d6f15893e2ed21a · report
ffc42bc6a89b86d8 · report
4a14389065cd27db · report
5a7d8b60a457be7b · report
b7e9b538a3a6d6c6 · report
ca5277eb2bdc881d · report
ebad0ef33daaaf54 · report
9a180909c46a9282 · report
7ac19ef8c80ec33c · report
25f32f40bc0cba2f · report
8c1b6a491ac9c94d · report
ffa8fed7bfc2815e · report
fe7e8a0ae28d05e2 · report
bfb7784ec2b95d72 · report
0b7afeadea0cacd7 · report
123fe9f28114f0d9 · report
96c97809b6f786f8 · report
9568d9f06a61b755 · report
4c4f887dd3b9d98b · report
f9b26425cdab4292 · report
4b62772e1596991e · report
decaab6917f91afb · report
4870ba968478dbba · report
5959383c642fb36b · report
31b966658e6d86f4 · report
3f76717dcef74a8c · report
f08bc60adbf978d5 · report
7479b68a714bcd2f · report
4587b69d57d6511b · report
738456e1b7388e7e · report
96a9939953d5992d · report
880696136d0e502a · report
ec127b7428b0c8df · report
62cf5e7db26b8795 · report
ae4ba8951ba45247 · report
17f8a36e011e8b2e · report
4f4fb90d76e06cbb · report
cdf3547dcf481932 · report
66dd0df8878140cf · report
f453b9c5d55c7987 · report
4f41d61541b9dd5c · report
d0396d59a24bd5e8 · report
8d69e6b197a5e237 · report
90253473da3f9ce0 · report
9c1fce3da0c14a5d · report
881893a4dbac5138 · report
0a4d54f8f5fc4644 · report
8de014c0f727cf2a · report
d3fdee80a50ec1cb · report
509c44b82896e66d · report
c2a4c8b84fb99859 · report
6110eed61afc93f5 · report
99f47e85886acbbe · report
f063e233528679dc · report
20f4e14e12340765 · report
f89f0dbcf985afec · report
5c9e2f1b8c51a5ad · report
18284ed4b2834099 · report
155ce11ed9c1f8d4 · report
fca099b402a99571 · report
ff92ad8f00e93bbc · report
1a3fa3dda9a8b100 · report
78d482d2a7373f3c · report
2e0088f9f6c77ccd · report
94cad4bc0776217e · report
ba28b6ff72c46019 · report
f6caf2197aa6a134 · report
9bca95aee23d1241 · report
a10bcfdf3e0502cc · report
b78a5e60fc79a562 · report
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Classification | InDL | Darknet53 | Average Recall | 88.53% | #7 of 9 | Archive leaderboard | report |
| Object Detection | COCO-O | YOLOv3 (DarkNet-53) | Average mAP | 14.8 | #41 of 45 | Archive leaderboard | report |
| Object Detection | COCO-O | YOLOv3 (DarkNet-53) | Effective Robustness | -0.37 | #41 of 45 | Archive leaderboard | report |
| Pedestrian Detection | DVTOD | YOLOv3 (Thermal) | mAP | 82.7 | #3 of 8 | Archive leaderboard | report |
| Pedestrian Detection | DVTOD | YOLOv3 (Visible) | mAP | 34.5 | #6 of 8 | Archive leaderboard | report |
| Real-Time Object Detection | COCO (Common Objects in Context) | YOLOv3-L | box AP | 33.0 | #80 of 82 | Archive leaderboard | report |
| Robust Object Detection | Cityscapes | Photometric distortion | mPC [AP] | 16.9 | #10 of 13 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Introduced by this paper: Fast-YOLOv3, YOLOv3, YOLOv8
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections