Papers › RTMDet: An Empirical Study of Designing Real-Time Object Detectors

RTMDet: An Empirical Study of Designing Real-Time Object Detectors

14 Dec 2022arXiv:2212.07784archive 2025-07-28

Chengqi Lyu, Wenwei Zhang, Haian Huang, Yue Zhou, Yudong Wang, Yanyi Liu, Shilong Zhang, Kai Chen

In this paper, we aim to design an efficient real-time object detector that exceeds the YOLO series and is easily extensible for many object recognition tasks such as instance segmentation and rotated object detection. To obtain a more efficient model architecture, we explore an architecture that has compatible capacities in the backbone and neck, constructed by a basic building block that consists of large-kernel depth-wise convolutions. We further introduce soft labels when calculating matching costs in the dynamic label assignment to improve accuracy. Together with better training techniques, the resulting object detector, named RTMDet, achieves 52.8% AP on COCO with 300+ FPS on an NVIDIA 3090 GPU, outperforming the current mainstream industrial detectors. RTMDet achieves the best parameter-accuracy trade-off with tiny/small/medium/large/extra-large model sizes for various application scenarios, and obtains new state-of-the-art performance on real-time instance segmentation and rotated object detection. We hope the experimental results can provide new insights into designing versatile real-time object detectors for many object recognition tasks. Code and models are released at https://github.com/open-mmlab/mmdetection/tree/3.x/configs/rtmdet.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

No sample from the archive's official repository (open-mmlab/mmdetection) is linked to this paper; the 20 samples below come from 3 community repositories. Syntology harvested code from open-mmlab/mmdetection for another paper.

open-mmlab/mmdetection: Syntology's graph links these samples to An Open and Comprehensive Pipeline for Unified Object Grounding and Detection (arXiv:2401.02361; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method.

Syntology 20 code samples harvested from 3 repositories linked to this paper: 16 ran (of which 0 constructed an object rather than computing a result; 15 with no instrument failure: 0 honoured, 0 violated, 15 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 20 samples from 3 repositories, 16 ran (15 with no instrument failure; 1 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 2 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-24 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

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

open-mmlab/mmdetection officialpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to An Open and Comprehensive Pipeline for Unified Object Grounding and Detection (arXiv:2401.02361; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
HVision-NKU/Strip-R-CNN mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
RangiLyu/mmdetection_test mentioned on GitHubpytorchApache-2.0Syntology: 7 samples, 5 ran (1 instrument-failed) · 2 unverified report
V3Det/mmdetection-V3Det mentioned on GitHubpytorchApache-2.0Syntology: not harvested report
cszzshi/SimD mentioned on GitHubpytorchApache-2.0Syntology: not harvested report
fiveai/MoCaE mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to MoCaE: Mixture of Calibrated Experts Significantly Improves Object Detection (arXiv:2309.14976; that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
open-edge-platform/geti mentioned on GitHubpytorchApache-2.0Syntology: not harvested report
open-mmlab/mmrotate mentioned on GitHubpytorchApache-2.0Syntology: 10 samples, 8 ran (0 instrument-failed) · 2 unverified report
open-mmlab/mmyolo mentioned on GitHubpytorchGPL-3.0Syntology: not harvested report
yuyi1005/point2rbox-mmrotate mentioned on GitHubpytorchApache-2.0Syntology: 3 samples, 3 ran (0 instrument-failed) · 0 unverified report
yxb-nku/strip-r-cnn mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
CycloneBoy/PPDetectionPytorch pytorchSyntology: not harvested report
PaddlePaddle/PaddleYOLO paddleGPL-3.0Syntology: not harvested report
open-edge-platform/training_extensions pytorchApache-2.0Syntology: not harvested 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

20 samples harvested: 16 ran (of which 0 constructed an object rather than computing a result; 15 with no instrument failure: 0 honoured, 0 violated, 15 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · our draft was wrong
15ran
4unverified

Licence: 3 of the 20 samples are pointer only, meaning Syntology does not serve that copy's text. For 3 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. 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; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

RangiLyu/mmdetection_test community (archive-listed) 7 samples, 5 ran (1 instrument-failed) · 2 unverified
check_link in 10 papers .dev_scripts/check_links.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 4e2dcdfe578946a0 · report fixture: signature match_tuple:MatchTuple/1/str, http_session:callable/0/?, logger:Logger/0/? · returned tuple · level recorded 2026-09-24
check_url in 17 papers .dev_scripts/check_links.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · cc93fdc812eca8f7 · report fixture: signature match_tuple:MatchTuple/1/str, http_session:callable/0/? · returned tuple · level recorded 2026-09-24
determine_gpus .dev_scripts/convert_train_benchmark_script.py ran fingerprinted Apache-2.0 (permissive) · fdef7359919a4918 · report fixture: signature cfg_name:str/0/str · returned tuple · level recorded 2026-09-25
process_checkpoint in 87 papers .dev_scripts/gather_models.py ran Apache-2.0 (permissive) · 138e112e22b0314e · report fixture: signature in_file:str/0/str, out_file:str/0/str · returned str · level recorded 2026-09-25
process_model_info in 21 papers .dev_scripts/convert_test_benchmark_script.py ran Apache-2.0 (permissive) · 3175ca3b1ec8c7b9 · report fixture: signature model_info:dict/0/?, work_dir:str/0/str · returned dict · level recorded 2026-09-24
check_path in 20 papers .dev_scripts/check_links.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · fd08138d4bde9cfa · report
ordered_yaml_dump in 33 papers .dev_scripts/gather_models.py unverified Apache-2.0 (permissive) · 8ff99a6bd2ebf598 · report
open-mmlab/mmrotate community (archive-listed) 10 samples, 8 ran (0 instrument-failed) · 2 unverified
bbox_flip mmrotate/core/bbox/transforms.py ran Apache-2.0 (permissive) · 376185dc39be213f · report fixture: signature bboxes:Tensor/2/float, img_shape:tuple/1/int, direction:str/0/str · returned Tensor · level recorded 2026-09-25
get_cls_results mmrotate/core/evaluation/eval_map.py ran Apache-2.0 (permissive) · d5e36bcdd9247df7 · report fixture: signature det_results:list/1/list, annotations:list/1/dict, class_id:int/0/int · returned tuple · level recorded 2026-09-25
get_multiscale_patch mmrotate/core/patch/split.py ran Apache-2.0 (permissive) · 01cb8ca0622cd942 · report fixture: signature sizes:list/1/int, steps:list/1/int, ratios:list/1/float · returned tuple · level recorded 2026-09-25
map_masks mmrotate/core/patch/merge_results.py ran Apache-2.0 (permissive) · 101fcc7bda94ac46 · report fixture: signature masks:list/1/ndarray, offset:ndarray/1/int, new_shape:tuple/1/int · returned list · level recorded 2026-09-25
rbbox2result mmrotate/core/bbox/transforms.py ran Apache-2.0 (permissive) · ce792faad2a6fa37 · report fixture: signature bboxes:Tensor/2/float, labels:Tensor/1/int, num_classes:int/0/int · returned list · level recorded 2026-09-25
rotated_anchor_inside_flags mmrotate/core/anchor/utils.py ran Apache-2.0 (permissive) · 0ea25f06be11d42b · report fixture: signature flat_anchors:Tensor/2/float, valid_flags:Tensor/1/bool, img_shape:tuple/1/int, allowed_border:int/0/int · returned Tensor · level recorded 2026-09-25
slide_window mmrotate/core/patch/split.py ran Apache-2.0 (permissive) · 4a5c249cf82b4cba · report fixture: signature width:int/0/int, height:int/0/int, sizes:list/1/int, steps:list/1/int, img_rate_thr:float/0/float · returned ndarray · level recorded 2026-09-25
translate_bboxes mmrotate/core/patch/merge_results.py ran Apache-2.0 (permissive) · 743d0cc9f86f379e · report fixture: signature bboxes:ndarray/2/float, offset:ndarray/1/float · returned ndarray · level recorded 2026-09-25
bbox_mapping_back mmrotate/core/bbox/transforms.py unverified Apache-2.0 (permissive) · bc4459d2ed8a1d71 · report
draw_rbboxes mmrotate/core/visualization/image.py unverified Apache-2.0 (permissive) · 9432da2297676e37 · report
yuyi1005/point2rbox-mmrotate community (archive-listed) 3 samples, 3 ran (0 instrument-failed) · 0 unverified
get_pattern_fill mmrotate/models/task_modules/synthesis_generators/point2rbox_generator.py ran fingerprinted Apache-2.0 (permissive) · ede6e43c8f430f7c · report fixture: signature w:int/0/int, h:int/0/int · returned ndarray · level recorded 2026-09-25
get_pattern_line mmrotate/models/task_modules/synthesis_generators/point2rbox_generator.py ran fingerprinted Apache-2.0 (permissive) · 45a05e5326bdf549 · report fixture: signature w:int/0/int, h:int/0/int · returned Tensor · level recorded 2026-09-25
get_pattern_rose mmrotate/models/task_modules/synthesis_generators/point2rbox_generator.py ran fingerprinted Apache-2.0 (permissive) · 1ed78f93e2bc1317 · report fixture: signature w:int/0/int, h:int/0/int · returned Tensor · level recorded 2026-09-25

Tasks

Instance SegmentationObjectObject DetectionObject Detection In Aerial ImagesObject RecognitionOne-stage Anchor-free Oriented Object DetectionOriented Object DetectionReal-Time Object DetectionReal-time Instance SegmentationSemantic Segmentationobject-detection

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Object Detection In Aerial Images DOTA 1.0 RTMDet-R-l mAP 81.33% #1 of 2 Archive leaderboard report
Object Detection In Aerial Images DOTA 1.0 RTMDet-R-l (single scale) mAP 80.16% #2 of 2 Archive leaderboard report
Object Detection In Aerial Images HRSC2016 RTMDet-R-tiny mAP-07 90.6 #5 of 9 Archive leaderboard report
Object Detection In Aerial Images HRSC2016 RTMDet-R-tiny mAP-12 97.10 #5 of 9 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) RTMDet box AP 52.8 #38 of 82 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-x AP50 67.4 #1 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-x AP75 47.8 #1 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-x APL 65.5 #1 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-x APS 22.2 #1 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-x Frame (fps) 188 (RTX3090) #1 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-x mask AP 44.6 #1 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-l AP50 66 #2 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-l AP75 47.0 #2 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-l APL 64.8 #2 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-l APM 48.0 #2 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-l APS 20.8 #2 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-l Frame (fps) 271 (RTX3090) #2 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-l mask AP 43.7 #2 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-m AP50 63.9 #3 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-m AP75 45.1 #3 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-m APL 63.1 #3 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-m APM 46.4 #3 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-m APS 19.3 #3 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-m Frame (fps) 371 (RTX3090) #3 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-m mask AP 42.1 #3 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-s AP50 59.3 #22 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-s AP75 41.3 #22 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-s APL 60.3 #22 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-s APM 42.3 #22 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-s APS 15.1 #22 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO RTMDet-Ins-s Frame (fps) 518 (RTX3090) #22 of 22 Archive leaderboard report
Real-time Instance Segmentation MSCOCO-1k RTMDet-Ins-x APM 49.0 #1 of 1 Archive leaderboard report
Real-time Instance Segmentation multi30k_test_2017_mscoco RTMDet-Ins-s mask AP 38.7 #1 of 1 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

RTMDet

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