Papers › D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement

D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement

17 Oct 2024arXiv:2410.13842archive 2025-07-28

Yansong Peng, Hebei Li, Peixi Wu, Yueyi Zhang, Xiaoyan Sun, Feng Wu

We introduce D-FINE, a powerful real-time object detector that achieves outstanding localization precision by redefining the bounding box regression task in DETR models. D-FINE comprises two key components: Fine-grained Distribution Refinement (FDR) and Global Optimal Localization Self-Distillation (GO-LSD). FDR transforms the regression process from predicting fixed coordinates to iteratively refining probability distributions, providing a fine-grained intermediate representation that significantly enhances localization accuracy. GO-LSD is a bidirectional optimization strategy that transfers localization knowledge from refined distributions to shallower layers through self-distillation, while also simplifying the residual prediction tasks for deeper layers. Additionally, D-FINE incorporates lightweight optimizations in computationally intensive modules and operations, achieving a better balance between speed and accuracy. Specifically, D-FINE-L / X achieves 54.0% / 55.8% AP on the COCO dataset at 124 / 78 FPS on an NVIDIA T4 GPU. When pretrained on Objects365, D-FINE-L / X attains 57.1% / 59.3% AP, surpassing all existing real-time detectors. Furthermore, our method significantly enhances the performance of a wide range of DETR models by up to 5.3% AP with negligible extra parameters and training costs. Our code and pretrained models: https://github.com/Peterande/D-FINE.

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

Syntology 14 code samples harvested from 1 repository linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 5 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 14 samples from 1 repository, 10 ran (5 with no instrument failure; 5 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 on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

Peterande/D-FINE officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 14 samples, 10 ran (5 instrument-failed) · 4 unverified report
open-edge-platform/geti mentioned on GitHubpytorchApache-2.0Syntology: not harvested report
shihuahuang95/deim mentioned on GitHubpytorchNOASSERTIONSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to DEIM: DETR with Improved Matching for Fast Convergence (arXiv:2412.04234; that paper's own run record: 9 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 4 where Syntology's instrument failed) · 4 unverified); nothing checks that they implement this paper's method. report
huggingface/transformers pytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Attention Is All You Need (arXiv:1706.03762; that paper's own run record: 610 ran (of which 293 constructed an object rather than computing a result; 529 with no instrument failure: 45 honoured, 15 violated, 469 with no contract checked; 81 where Syntology's instrument failed) · 336 unverified), HuggingFace's Transformers: State-of-the-art Natural Language Processing (arXiv:1910.03771; that paper's own run record: 3 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; 3 where Syntology's instrument failed) · 0 unverified), DeBERTa: Decoding-enhanced BERT with Disentangled Attention (arXiv:2006.03654; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 2 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 9 unverified) and 1 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. 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

14 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 5 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.

3ran · our draft was wrong
2ran; the contract check could not run
5ran
4unverified

Licence: 0 of the 14 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 Peterande/D-FINE. “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.

Peterande/D-FINE official repository 14 samples, 10 ran (5 instrument-failed) · 4 unverified
dictify in 33 papers src/core/yaml_utils.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 55ac89cb59e7fe93 · report fixture: signature s:str/0/str, v:int/0/int · returned dict · level recorded 2026-09-24
elementwise_box_iou src/misc/box_ops.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · f032fa04460c4fc9 · report fixture: signature boxes1:Tensor/2/float, boxes2:Tensor/2/float · returned tuple · level recorded 2026-09-24
elementwise_generalized_box_iou src/misc/box_ops.py ran; the contract check could not run (RAISES) fingerprinted Apache-2.0 (permissive) · 96986c2e89b1597c · report fixture: signature boxes1:Tensor/2/float, boxes2:Tensor/2/float · returned Tensor · level recorded 2026-09-24
extract_schema src/core/workspace.py ran · our draft was wrong Apache-2.0 (permissive) · 1202e5e128785814 · report fixture: returned dict · level recorded 2026-09-24
generalized_box_iou in 10 papers src/misc/box_ops.py ran; the contract check could not run (RAISES) fingerprinted Apache-2.0 (permissive) · 2f52077940b53c42 · report fixture: signature boxes1:Tensor/2/float, boxes2:Tensor/2/float · returned Tensor · level recorded 2026-09-24
load_config in 389 papers src/core/yaml_utils.py ran Apache-2.0 (permissive) · 024f9ddd58b25511 · report fixture: signature file_path:str/0/str, cfg:dict/0/? · returned dict · level recorded 2026-09-24
merge_dict in 13 papers src/core/yaml_utils.py ran Apache-2.0 (permissive) · 288efb4f48e3358a · report fixture: signature dct:dict/0/?, another_dct:dict/0/?, inplace:bool/0/bool · returned dict · level recorded 2026-09-24
register in 66 papers src/core/workspace.py ran Apache-2.0 (permissive) · 2a50c87f96494841 · report fixture: signature dct:dict/0/?, name:NoneType/0/?, force:bool/0/bool · returned function · level recorded 2026-09-24
setup_distributed in 11 papers src/misc/dist_utils.py ran Apache-2.0 (permissive) · a65eab71a65df60e · report fixture: returned bool · level recorded 2026-09-24
warp_model src/misc/dist_utils.py ran Apache-2.0 (permissive) · 18349a0bf298c043 · report fixture: signature model:callable/0/?, sync_bn:bool/0/bool, dist_mode:str/0/str, find_unused_parameters:bool/0/bool, compile:bool/0/bool, compile_mode:str/0/str · returned SimpleModel · level recorded 2026-09-24
all_gather in 119 papers src/misc/logger.py unverified Apache-2.0 (permissive) · 0216311af005f21f · report
create in 17 papers src/core/workspace.py unverified Apache-2.0 (permissive) · eb5e01d60557539c · report
de_model src/misc/dist_utils.py unverified Apache-2.0 (permissive) · 8c4ff9fb866ea191 · report
reduce_dict in 73 papers src/misc/logger.py unverified Apache-2.0 (permissive) · e6c2f1b16c9d4813 · report

Tasks

Real-Time Object Detectionregression

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) D-FINE-X+ FPS (V100, b=1) 78 (T4) #2 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-X+ box AP 59.3 #2 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-X FPS (V100, b=1) 78 (T4) #11 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-X box AP 55.8 #11 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-M+ FPS (V100, b=1) 178 (T4) #15 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-M+ box AP 55.1 #15 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-L FPS (V100, b=1) 124 (T4) #26 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-L box AP 54.0 #26 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-M FPS (V100, b=1) 178 (T4) #45 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-M box AP 52.3 #45 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-S+ FPS (V100, b=1) 287 (T4) #54 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-S+ box AP 50.7 #54 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-S FPS (V100, b=1) 287 (T4) #61 of 82 Archive leaderboard report
Real-Time Object Detection COCO (Common Objects in Context) D-FINE-S box AP 48.5 #61 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 ConnectionSPEEDSoftmaxTransformer

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