Papers › DEIM: DETR with Improved Matching for Fast Convergence

DEIM: DETR with Improved Matching for Fast Convergence

5 Dec 2024CVPR 2025 1arXiv:2412.04234archive 2025-07-28

Shihua Huang, Zhichao Lu, Xiaodong Cun, Yongjun Yu, Xiao Zhou, Xi Shen

We introduce DEIM, an innovative and efficient training framework designed to accelerate convergence in real-time object detection with Transformer-based architectures (DETR). To mitigate the sparse supervision inherent in one-to-one (O2O) matching in DETR models, DEIM employs a Dense O2O matching strategy. This approach increases the number of positive samples per image by incorporating additional targets, using standard data augmentation techniques. While Dense O2O matching speeds up convergence, it also introduces numerous low-quality matches that could affect performance. To address this, we propose the Matchability-Aware Loss (MAL), a novel loss function that optimizes matches across various quality levels, enhancing the effectiveness of Dense O2O. Extensive experiments on the COCO dataset validate the efficacy of DEIM. When integrated with RT-DETR and D-FINE, it consistently boosts performance while reducing training time by 50%. Notably, paired with RT-DETRv2, DEIM achieves 53.2% AP in a single day of training on an NVIDIA 4090 GPU. Additionally, DEIM-trained real-time models outperform leading real-time object detectors, with DEIM-D-FINE-L and DEIM-D-FINE-X achieving 54.7% and 56.5% AP at 124 and 78 FPS on an NVIDIA T4 GPU, respectively, without the need for additional data. We believe DEIM sets a new baseline for advancements in real-time object detection. Our code and pre-trained models are available at https://github.com/ShihuaHuang95/DEIM.

PaperPDFConference PDFCodeCode 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="2412.04234")

Code

Syntology Ran 1 of 13 code samples harvested from 1 repository linked to this paper; 12 have no recorded run. Of those that ran: 1 ran · honoured contract.

By repository: official repository: 13 samples from 1 repository, 1 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

shihuahuang95/deim officialmentioned in papermentioned on GitHubpytorchNOASSERTION 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

13 samples harvested; 1 ran; 1 honoured the contract we drafted; 12 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 · honoured contract
12unverified

Licence: 13 of the 13 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 shihuahuang95/deim. “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.

autopad shihuahuang95/deim/engine/backbone/csp_darknet.py official repository ran · honoured contract no licence file found · pointer only · 988a3c854b1b13d0 · report
bias_init_with_prob shihuahuang95/deim/engine/deim/utils.py official repository unverified licence not identified · pointer only · 8c4a2df028bbf447 · report
box_cxcywh_to_xyxy shihuahuang95/deim/engine/deim/box_ops.py official repository unverified licence not identified · pointer only · cb48d7481c17a1f7 · report
box_iou shihuahuang95/deim/engine/deim/box_ops.py official repository unverified licence not identified · pointer only · 253961917cbe7713 · report
box_xyxy_to_cxcywh shihuahuang95/deim/engine/deim/box_ops.py official repository unverified licence not identified · pointer only · db44d81a87192d0f · report
deformable_attention_core_func shihuahuang95/deim/engine/deim/utils.py official repository unverified licence not identified · pointer only · 9524bdc261cbc2dc · report
distance2bbox shihuahuang95/deim/engine/deim/dfine_utils.py official repository unverified licence not identified · pointer only · 07ceddb7b30df7b3 · report
get_contrastive_denoising_training_group shihuahuang95/deim/engine/deim/denoising.py official repository unverified licence not identified · pointer only · 4e994353f1d6e99a · report
inverse_sigmoid shihuahuang95/deim/engine/deim/utils.py official repository unverified licence not identified · pointer only · 2fff86fc7c917b0c · report
make_divisible shihuahuang95/deim/engine/backbone/csp_darknet.py official repository unverified licence not identified · pointer only · 2a08c4c250501510 · report
mod shihuahuang95/deim/engine/deim/postprocessor.py official repository unverified licence not identified · pointer only · ec4284e2142c46f9 · report
translate_gt shihuahuang95/deim/engine/deim/dfine_utils.py official repository unverified licence not identified · pointer only · 0f787f3e8b97aba1 · report
weighting_function shihuahuang95/deim/engine/deim/dfine_utils.py official repository unverified licence not identified · pointer only · 1a45b1a466ee92a4 · report

Tasks

Data AugmentationObject DetectionReal-Time Object Detectionobject-detection

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Real-Time Object Detection COCO (Common Objects in Context) DEIM-D-FINE-X+ FPS (V100, b=1) 78 (T4) #1 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) DEIM-D-FINE-X+ box AP 59.5 #1 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) DEIM-D-FINE-X FPS (V100, b=1) 78 (T4) #8 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) DEIM-D-FINE-X box AP 56.5 #8 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) DEIM-D-FINE-L FPS (V100, b=1) 124 (T4) #20 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) DEIM-D-FINE-L box AP 54.7 #20 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) DEIM-D-FINE-M FPS (V100, b=1) 178 (T4) #39 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) DEIM-D-FINE-M box AP 52.7 #39 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) DEIM-D-FINE-S FPS (V100, b=1) 287 (T4) #60 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) DEIM-D-FINE-S box AP 49.0 #60 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) DEIM-D-FINE-X+ FPS (V100, b=1) 78 (T4) #82 of 82 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

Absolute Position EncodingsAdamAttentionBPEConvolutionDense ConnectionsDetrDropoutFeedforward NetworkLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

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