Papers › Dynamic Head: Unifying Object Detection Heads with Attentions
Dynamic Head: Unifying Object Detection Heads with Attentions
Xiyang Dai, Yinpeng Chen, Bin Xiao, Dongdong Chen, Mengchen Liu, Lu Yuan, Lei Zhang
The complex nature of combining localization and classification in object detection has resulted in the flourished development of methods. Previous works tried to improve the performance in various object detection heads but failed to present a unified view. In this paper, we present a novel dynamic head framework to unify object detection heads with attentions. By coherently combining multiple self-attention mechanisms between feature levels for scale-awareness, among spatial locations for spatial-awareness, and within output channels for task-awareness, the proposed approach significantly improves the representation ability of object detection heads without any computational overhead. Further experiments demonstrate that the effectiveness and efficiency of the proposed dynamic head on the COCO benchmark. With a standard ResNeXt-101-DCN backbone, we largely improve the performance over popular object detectors and achieve a new state-of-the-art at 54.0 AP. Furthermore, with latest transformer backbone and extra data, we can push current best COCO result to a new record at 60.6 AP. The code will be released at https://github.com/microsoft/DynamicHead.
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="2106.08322")
Code
Syntology Ran 7 of 9 code samples harvested from 2 repositories linked to this paper; 2 have no recorded run. Of those that ran: 1 ran · our draft was wrong; 1 ran · fixture could not drive it; 5 ran with no contract checked.
By repository: official repository: 3 samples from 1 repository, 2 ran; community (archive-listed): 6 samples from 1 repository, 5 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.
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
9 samples harvested; 7 ran; 0 honoured the contract we drafted; 2 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: 0 of the 9 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 2 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.
f9fd6241d935f07b · report
fb32094c6dbece71 · report
2028ba00ac3a78ae · report
43a848f530c2b68c · report
f1f13347d3ea0492 · report
ae71a995161d732d · report
0d6fb3b73f74caa3 · report
4c367c8f60b22fe0 · report
0b96184174adb44f · report
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Object Detection | COCO 2017 val | DyHead (Swin-T, multi scale) | AP50 | 68 | #33 of 33 | Archive leaderboard | report |
| Object Detection | COCO 2017 val | DyHead (Swin-T, multi scale) | AP75 | 54.3 | #33 of 33 | Archive leaderboard | report |
| Object Detection | COCO 2017 val | DyHead (Swin-T, multi scale) | APL | 64.2 | #33 of 33 | Archive leaderboard | report |
| Object Detection | COCO minival | DyHead (Swin-L, multi scale, self-training) | AP50 | 78.2 | #25 of 220 | Archive leaderboard | report |
| Object Detection | COCO minival | DyHead (Swin-L, multi scale, self-training) | APL | 74.2 | #25 of 220 | Archive leaderboard | report |
| Object Detection | COCO minival | DyHead (Swin-L, multi scale, self-training) | box AP | 60.3 | #25 of 220 | Archive leaderboard | report |
| Object Detection | COCO minival | DyHead (Swin-L, multi scale) | AP50 | 76.8 | #36 of 220 | Archive leaderboard | report |
| Object Detection | COCO minival | DyHead (Swin-L, multi scale) | APL | 73.2 | #36 of 220 | Archive leaderboard | report |
| Object Detection | COCO minival | DyHead (Swin-L, multi scale) | APM | 62.2 | #36 of 220 | Archive leaderboard | report |
| Object Detection | COCO minival | DyHead (Swin-L, multi scale) | APS | 44.5 | #36 of 220 | Archive leaderboard | report |
| Object Detection | COCO minival | DyHead (Swin-L, multi scale) | box AP | 58.4 | #36 of 220 | Archive leaderboard | report |
| Object Detection | COCO minival | DyHead (ResNet-101) | box AP | 46.5 | #103 of 220 | Archive leaderboard | report |
| Object Detection | COCO minival | DyHead (ResNeXt-64x4d-101-DCN, multi scale) | APL | 66.3 | #220 of 220 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (Swin-L, multi scale, self-training) | AP50 | 78.5 | #26 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (Swin-L, multi scale, self-training) | AP75 | 66.6 | #26 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (Swin-L, multi scale, self-training) | APL | 74.2 | #26 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (Swin-L, multi scale, self-training) | APM | 64.0 | #26 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (Swin-L, multi scale, self-training) | box mAP | 60.6 | #26 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (Swin-L, multi scale) | AP50 | 77.1 | #34 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (Swin-L, multi scale) | AP75 | 64.5 | #34 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (Swin-L, multi scale) | APL | 72.8 | #34 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (Swin-L, multi scale) | APM | 62.0 | #34 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (Swin-L, multi scale) | box mAP | 58.7 | #34 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (ResNeXt-64x4d-101-DCN, multi scale) | AP50 | 72.1 | #58 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (ResNeXt-64x4d-101-DCN, multi scale) | AP75 | 59.3 | #58 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (ResNeXt-64x4d-101-DCN, multi scale) | box mAP | 54 | #58 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (ResNeXt-64x4d-101) | AP50 | 65.7 | #116 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (ResNeXt-64x4d-101) | AP75 | 51.9 | #116 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (ResNeXt-64x4d-101) | box mAP | 47.7 | #116 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (ResNet-50) | AP50 | 60.7 | #165 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (ResNet-50) | AP75 | 46.8 | #165 of 225 | Archive leaderboard | report |
| Object Detection | COCO test-dev | DyHead (ResNet-50) | box mAP | 43 | #165 of 225 | Archive leaderboard | report |
| Object Detection | COCO-O | DyHead (Swin-L) | Average mAP | 35.3 | #9 of 45 | Archive leaderboard | report |
| Object Detection | COCO-O | DyHead (Swin-L) | Effective Robustness | 10.00 | #9 of 45 | Archive leaderboard | report |
| Object Detection | COCO-O | DyHead (ResNet-50) | Average mAP | 19.3 | #31 of 45 | Archive leaderboard | report |
| Object Detection | COCO-O | DyHead (ResNet-50) | Effective Robustness | 0.16 | #31 of 45 | 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