Papers › DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection

DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection

7 Mar 2022arXiv:2203.03605archive 2025-07-28

Hao Zhang, Feng Li, Shilong Liu, Lei Zhang, Hang Su, Jun Zhu, Lionel M. Ni, Heung-Yeung Shum

We present DINO (\textbf{D}ETR with \textbf{I}mproved de\textbf{N}oising anch\textbf{O}r boxes), a state-of-the-art end-to-end object detector. % in this paper. DINO improves over previous DETR-like models in performance and efficiency by using a contrastive way for denoising training, a mixed query selection method for anchor initialization, and a look forward twice scheme for box prediction. DINO achieves $49.4$AP in $12$ epochs and $51.3$AP in $24$ epochs on COCO with a ResNet-50 backbone and multi-scale features, yielding a significant improvement of +6.0\textbf{AP} and +2.7\textbf{AP}, respectively, compared to DN-DETR, the previous best DETR-like model. DINO scales well in both model size and data size. Without bells and whistles, after pre-training on the Objects365 dataset with a SwinL backbone, DINO obtains the best results on both COCO \texttt{val2017} (63.2\textbf{AP}) and \texttt{test-dev} (\textbf{63.3AP}). Compared to other models on the leaderboard, DINO significantly reduces its model size and pre-training data size while achieving better results. Our code will be available at \url{https://github.com/IDEACVR/DINO}.

PaperPDFCodeCode Syntology ran

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="2203.03605")

Code

Syntology Ran 7 of 15 code samples harvested from 3 repositories linked to this paper; 8 have no recorded run. Of those that ran: 1 ran · violated contract; 2 ran · our draft was wrong; 2 ran · fixture could not drive it; 2 ran with no contract checked.

By repository: community (archive-listed): 15 samples from 3 repositories, 7 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

16 repositories listed; official and paper-mentioned ones first.

IDEACVR/DINO officialmentioned in papermentioned on GitHubpytorchApache-2.0 report
IDEA-Research/detrex mentioned on GitHubpytorch report
IDEA-opensource/DAB-DETR mentioned on GitHubpytorch report
IDEA-opensource/DN-DETR mentioned on GitHubpytorchApache-2.0 report
IDEACVR/MaskDINO mentioned on GitHubpytorchApache-2.0 report
NVlabs/FasterViT mentioned on GitHubpytorch report
horrible-dong/teamdetr mentioned on GitHubpytorchApache-2.0 report
idea-research/dab-detr mentioned on GitHubpytorchApache-2.0 report
idea-research/dino mentioned on GitHubpytorchApache-2.0 report
idea-research/dn-detr mentioned on GitHubpytorch report
idea-research/maskdino mentioned on GitHubpytorchApache-2.0 report
lucasjinreal/yolov7_d2 mentioned on GitHubpytorchGPL-3.0 report

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

15 samples harvested; 7 ran; 0 honoured the contract we drafted; 8 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.

1ran · violated contract
2ran · our draft was wrong
2ran · fixture could not drive it
2ran
8unverified

Licence: 5 of the 15 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 3 repositories linked to this paper, official or community; each sample names its own and says which. “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.

dice_loss idea-research/dino/models/dino/segmentation.py community (archive-listed) ran · violated contract fingerprinted Apache-2.0 recorded; this copy not marked cleared · pointer only · ac8fe530cdad4d8c · report
dn_post_process idea-research/dino/models/dino/dn_components.py community (archive-listed) ran Apache-2.0 recorded; this copy not marked cleared · pointer only · dc279ea790da466f · report
gen_encoder_output_proposals idea-research/dino/models/dino/utils.py community (archive-listed) ran Apache-2.0 recorded; this copy not marked cleared · pointer only · e3b2768e6a9c1cb5 · report
gen_sineembed_for_position idea-research/dino/models/dino/utils.py community (archive-listed) ran · fixture could not drive it fingerprinted Apache-2.0 recorded; this copy not marked cleared · pointer only · 4c82654ae6131ecc · report
sigmoid_focal_loss IDEACVR/MaskDINO/maskdino/modeling/criterion.py community (archive-listed) ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 5c0711aada67957e · report
window_partition IDEACVR/MaskDINO/maskdino/modeling/backbone/swin.py community (archive-listed) ran · fixture could not drive it fingerprinted Apache-2.0 (permissive) · f9fd6241d935f07b · report
window_reverse IDEACVR/MaskDINO/maskdino/modeling/backbone/swin.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · fb32094c6dbece71 · report
batch_dice_loss IDEACVR/MaskDINO/maskdino/modeling/matcher.py community (archive-listed) unverified Apache-2.0 (permissive) · bc2cb481a75c370d · report
batch_sigmoid_ce_loss IDEACVR/MaskDINO/maskdino/modeling/matcher.py community (archive-listed) unverified Apache-2.0 (permissive) · 1edd24985036b0bf · report
build_focalnet IDEACVR/MaskDINO/maskdino/modeling/backbone/focal.py community (archive-listed) unverified Apache-2.0 (permissive) · fb3c1f63d2222804 · report
ciou horrible-dong/teamdetr/Team-DAB-DETR/util/box_loss.py community (archive-listed) unverified Apache-2.0 (permissive) · 4373b99bf6b5fd7d · report
dice_loss IDEACVR/MaskDINO/maskdino/modeling/criterion.py community (archive-listed) unverified Apache-2.0 (permissive) · 89f75e54ff128be0 · report
diou horrible-dong/teamdetr/Team-DAB-DETR/util/box_loss.py community (archive-listed) unverified Apache-2.0 (permissive) · 7818a5040676702a · report
multi_head_attention_forward idea-research/dino/models/dino/attention.py community (archive-listed) unverified Apache-2.0 recorded; this copy not marked cleared · pointer only · a82091666b9db8cf · report
sigmoid_ce_loss IDEACVR/MaskDINO/maskdino/modeling/criterion.py community (archive-listed) unverified Apache-2.0 (permissive) · d0c61e8dba511aa3 · report

Tasks

Object DetectionReal-Time Object Detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Object Detection COCO minival DINO (Swin-L) box AP 63.2 #13 of 220 Archive leaderboard report
Object Detection COCO minival DINO-5scale (24 epoch) AP50 69.1 #73 of 220 Archive leaderboard report
Object Detection COCO minival DINO-5scale (24 epoch) AP75 56 #73 of 220 Archive leaderboard report
Object Detection COCO minival DINO-5scale (24 epoch) APL 65.8 #73 of 220 Archive leaderboard report
Object Detection COCO minival DINO-5scale (24 epoch) APM 54.2 #73 of 220 Archive leaderboard report
Object Detection COCO minival DINO-5scale (24 epoch) APS 34.5 #73 of 220 Archive leaderboard report
Object Detection COCO minival DINO-5scale (24 epoch) box AP 51.3 #73 of 220 Archive leaderboard report
Object Detection COCO minival DINO-5scale (36 epoch) AP50 69 #74 of 220 Archive leaderboard report
Object Detection COCO minival DINO-5scale (36 epoch) AP75 55.8 #74 of 220 Archive leaderboard report
Object Detection COCO minival DINO-5scale (36 epoch) APL 65.3 #74 of 220 Archive leaderboard report
Object Detection COCO minival DINO-5scale (36 epoch) APM 54.3 #74 of 220 Archive leaderboard report
Object Detection COCO minival DINO-5scale (36 epoch) APS 35 #74 of 220 Archive leaderboard report
Object Detection COCO minival DINO-5scale (36 epoch) box AP 51.2 #74 of 220 Archive leaderboard report
Object Detection COCO test-dev DINO (Swin-L,multi-scale, TTA) box mAP 63.3 #17 of 225 Archive leaderboard report
Object Detection COCO-O DINO (Swin-L) Average mAP 42.1 #5 of 45 Archive leaderboard report
Object Detection COCO-O DINO (Swin-L) Effective Robustness 15.76 #5 of 45 Archive leaderboard report
Object Detection SA-Det-100k DINO (ResNet50 1x VFL) AP 43.7 #2 of 2 Archive leaderboard report
Object Detection SA-Det-100k DINO (ResNet50 1x VFL) AP50 52.0 #2 of 2 Archive leaderboard report
Object Detection SA-Det-100k DINO (ResNet50 1x VFL) AP75 47.7 #2 of 2 Archive leaderboard report
Object Detection SA-Det-100k DINO (ResNet50 1x VFL) APL 61.5 #2 of 2 Archive leaderboard report
Object Detection SA-Det-100k DINO (ResNet50 1x VFL) APM 43.0 #2 of 2 Archive leaderboard report
Object Detection SA-Det-100k DINO (ResNet50 1x VFL) APS 5.8 #2 of 2 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.

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