Methods › General › Attention Mechanisms › Coordinate attention
Coordinate attention
Introduced by Qibin Hou et al. in Coordinate Attention for Efficient Mobile Network Design
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
Hou et al. proposed coordinate attention, a novel attention mechanism which embeds positional information into channel attention, so that the network can focus on large important regions at little computational cost.
The coordinate attention mechanism has two consecutive steps, coordinate information embedding and coordinate attention generation. First, two spatial extents of pooling kernels encode each channel horizontally and vertically. In the second step, a shared 1×1 convolutional transformation function is applied to the concatenated outputs of the two pooling layers. Then coordinate attention splits the resulting tensor into two separate tensors to yield attention vectors with the same number of channels for horizontal and vertical coordinates of the input X along. This can be written as zʰ = GAPʰ(X) zʷ = GAPʷ(X) f = δ(BN(Conv₁^(1×1)([zʰ;zʷ]))) fʰ, fʷ = Split(f) sʰ = σ(Convₕ^(1×1)(fʰ)) sʷ = σ(Conv_w^(1×1)(fʷ)) Y = X sʰ sʷ where GAPʰ and GAPʷ denote pooling functions for vertical and horizontal coordinates, and sʰ ∈ℝ^(C×1×W) and sʷ ∈ℝ^(C×H×1) represent corresponding attention weights.
Using coordinate attention, the network can accurately obtain the position of a targeted object. This approach has a larger receptive field than BAM and CBAM. Like an SE block, it also models cross-channel relationships, effectively enhancing the expressive power of the learned features. Due to its lightweight design and flexibility, it can be easily used in classical building blocks of mobile networks.
In Syntology repository known, no verified implementation (declared_code; repositories: 2 listed of 2 in the graph, strongest link: official). The lookup matched 'Coordinate Attention' ignoring case. This answer is about arXiv:2103.02907, the paper the archive credits with this method.
Code Syntology holds under other names, as the lookup lists them (code_nearby); not a complete list, and none of them is this method.
- RPC-Attention · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='RPC-Attention') - Slot Attention · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='Slot Attention') - Primal-Attention · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='Primal-Attention') - Alignment Attention · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='Alignment Attention') - Sequential Attention · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='Sequential Attention') - Elliptical Attention · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='Elliptical Attention')
Origin paper of this answer, arXiv:2103.02907 (its page on this site · its page on Syntology).
Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.
What each level means, quoted from Syntology's server card: V0 “imports”; V1 “executes against a synthesized fixture and returns a value of the recorded type”; V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” The card adds: “No rung on this ladder compares the served code against the authors' implementation, so treat every level as evidence about runnability and internal agreement, not about matching the paper's results.” A level quoted in two parts is one definition the card writes as two clauses joined by a double hyphen, each clause quoted on its own (how these levels are quoted).
Papers archive 2025-07-28
archive papers tagged: 31 · with a code link: 8 · where Syntology ran a sample: 1 (0 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument) Syntology
Show: all tagged papersonly where code ran (1 of 31 tagged: 0 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument)
Page 1 of 1: papers 1 to 31 of 31, newest first by the archive's date (ties by slug), in archive order.
Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.
-
Design description of Wisdom Computing Persperctive 2 May 2025 · 0 repositories · arXiv:2505.03800
-
Enhancing Traffic Sign Recognition On The Performance Based On Yolov8 2 Apr 2025 · 0 repositories · arXiv:2504.02884
-
LightEndoStereo: A Real-time Lightweight Stereo Matching Method for Endoscopy Images 2 Mar 2025 · 1 repository · arXiv:2503.00731
-
A Physics-Inspired Deep Learning Framework with Polar Coordinate Attention for Ptychographic Imaging 25 Nov 2024 · 1 repository · arXiv:2412.06806
-
Hyperspectral Imaging-Based Perception in Autonomous Driving Scenarios: Benchmarking Baseline Semantic Segmentation Models 29 Oct 2024 · 1 repository · arXiv:2410.22101
-
Optimizing YOLO Architectures for Optimal Road Damage Detection and Classification: A Comparative Study from YOLOv7 to YOLOv10 10 Oct 2024 · 0 repositories · arXiv:2410.08409
-
Improved Unet model for brain tumor image segmentation based on ASPP-coordinate attention mechanism 13 Sep 2024 · 0 repositories · arXiv:2409.08588
-
RICAU-Net: Residual-block Inspired Coordinate Attention U-Net for Segmentation of Small and Sparse Calcium Lesions in Cardiac CT 11 Sep 2024 · 0 repositories · arXiv:2409.06993
-
ALSS-YOLO: An Adaptive Lightweight Channel Split and Shuffling Network for TIR Wildlife Detection in UAV Imagery 10 Sep 2024 · 1 repository · arXiv:2409.06259
-
CSANet: Channel Spatial Attention Network for Robust 3D Face Alignment and Reconstruction 30 May 2024 · 0 repositories · arXiv:2405.19659
-
ELA: Efficient Local Attention for Deep Convolutional Neural Networks 2 Mar 2024 · 0 repositories · arXiv:2403.01123
-
Deep Linear Array Pushbroom Image Restoration: A Degradation Pipeline and Jitter-Aware Restoration Network 16 Jan 2024 · 1 repository · arXiv:2401.08171Syntology official (archive's flag): 5 ran · 5 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; 5 where Syntology's instrument failed) · 3 unverified (of 8 harvested samples)
-
YOLO algorithm with hybrid attention feature pyramid network for solder joint defect detection 2 Jan 2024 · 0 repositories · arXiv:2401.01214
-
YOLOv5s-BC: An improved YOLOv5s-based method for real-time apple detection 10 Nov 2023 · 0 repositories · arXiv:2311.05811
-
Marine Debris Detection in Satellite Surveillance using Attention Mechanisms 9 Jul 2023 · 0 repositories · arXiv:2307.04128
-
Multi-cropping Contrastive Learning and Domain Consistency for Unsupervised Image-to-Image Translation 24 Apr 2023 · 0 repositories · arXiv:2304.12235
-
Two-stage MR Image Segmentation Method for Brain Tumors based on Attention Mechanism 17 Apr 2023 · 0 repositories · arXiv:2304.08072
-
Fast vehicle detection algorithm based on lightweight YOLO7-tiny 12 Apr 2023 · 0 repositories · arXiv:2304.06002
-
PCCA-Model: an attention module for medical image segmentation 1 Apr 2023 · 0 repositories
-
TWR-MCAE: A Data Augmentation Method for Through-the-Wall Radar Human Motion Recognition 6 Jan 2023 · 0 repositories · arXiv:2301.02488
-
Road Damages Detection and Classification with YOLOv7 31 Oct 2022 · 0 repositories · arXiv:2211.00091
-
AIA: Attention in Attention Within Collaborate Domains 7 Oct 2022 · 1 repository
-
A New Method on Mask-Wearing Detection for Natural Population Based on Improved YOLOv4 24 Aug 2022 · 0 repositories · arXiv:2208.11353
-
Improved lightweight identification of agricultural diseases based on MobileNetV3 19 Jul 2022 · 0 repositories · arXiv:2207.11238
-
An Improved Lightweight YOLOv5 Model Based on Attention Mechanism for Face Mask Detection 30 Mar 2022 · 1 repository · arXiv:2203.16506
-
SAR-ShipNet: SAR-Ship Detection Neural Network via Bidirectional Coordinate Attention and Multi-resolution Feature Fusion 29 Mar 2022 · 0 repositories · arXiv:2203.15480
-
Improving Generalization of Deep Networks for Estimating Physical Properties of Containers and Fillings 2 Mar 2022 · 0 repositories · arXiv:2203.01192
-
A Lightweight Dual-Domain Attention Framework for Sparse-View CT Reconstruction 19 Feb 2022 · 0 repositories · arXiv:2202.09609
-
Attention-Guided NIR Image Colorization via Adaptive Fusion of Semantic and Texture Clues 20 Jul 2021 · 0 repositories · arXiv:2107.09237
-
Joint Attention for Multi-Agent Coordination and Social Learning 15 Apr 2021 · 0 repositories · arXiv:2104.07750
-
Coordinate Attention for Efficient Mobile Network Design 4 Mar 2021 · 2 repositories · arXiv:2103.02907
Tasks archive 2025-07-28
20 shown of 51 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
| Task | Papers |
|---|---|
| Semantic Segmentation | 6 |
| Data Augmentation | 4 |
| Object Detection | 4 |
| Image Segmentation | 3 |
| Segmentation | 3 |
| object-detection | 3 |
| Autonomous Driving | 2 |
| Deep Learning | 2 |
| Generative Adversarial Network | 2 |
| Mamba | 2 |
| Road Damage Detection | 2 |
| 3D Face Alignment | 1 |
| Autonomous Vehicles | 1 |
| Brain Tumor Segmentation | 1 |
| CT Reconstruction | 1 |
| Classification | 1 |
| Colorization | 1 |
| Computed Tomography (CT) | 1 |
| Contrastive Learning | 1 |
| Deep Attention | 1 |
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
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