Methods › General › Affinity Functions › Embedded Gaussian Affinity

Embedded Gaussian Affinity

8 papers tagged archive 2025-07-28

Introduced by Xiaolong Wang et al. in Non-local Neural Networks

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

Embedded Gaussian Affinity is a type of affinity or self-similarity function between two points 𝐱ᵢ and 𝐱ⱼ that uses a Gaussian function in an embedding space:

f(𝐱ᵢ, 𝐱ⱼ) = e^(θ(𝐱ᵢ)ᵀϕ(𝐱ⱼ))

Here θ(xᵢ) = W_θxᵢ and ϕ(xⱼ) = Wᵩxⱼ are two embeddings.

Note that the self-attention module used in the original Transformer model is a special case of non-local operations in the embedded Gaussian version. This can be seen from the fact that for a given i, 1/(𝒞(𝐱))∑_(∀j)f(𝐱ᵢ, 𝐱ⱼ)g(𝐱ⱼ) becomes the softmax computation along the dimension j. So we have 𝐲 = softmax(𝐱ᵀWᵀ_θW_ϕ𝐱)g(𝐱), which is the self-attention form in the Transformer model. This shows how we can relate this recent self-attention model to the classic computer vision method of non-local means.

PaperSourceSee Code · tea1528/Non-Local-NN-Pytorch

Papers archive 2025-07-28

8 shown of 8, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.

Tasks archive 2025-07-28

19 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

TaskPapers
Object Detection4
object-detection3
Action Classification1
Action Recognition1
Ensemble Learning1
Fire Detection1
Food recommendation1
Fracture detection1
Instance Segmentation1
Keypoint Detection1
Management1
Medical Object Detection1
Nutrition1
Object Localization1
Pose Estimation1
Position1
Text-To-SQL1
Transfer Learning1
Video Classification1

Usage over time archive 2025-07-28

Papers per year tagged with Embedded Gaussian Affinity: 2017 to 2024, peak 3 3 0 2017: 1 paper 2017 2018: 0 papers 2018 2019: 1 paper 2019 2020: 1 paper 2020 2021: 3 papers 2021 2022: 0 papers 2022 2023: 1 paper 2023 2024: 1 paper 2024
Papers per year the archive tags with this method, by the paper's archive date (8 dated). Bars are counts, not a trend claim.

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

Affinity Functions

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