Methods › Computer Vision › Image Model Blocks › Ghost Module

Ghost Module

30 papers tagged archive 2025-07-28

Introduced by Kai Han et al. in GhostNet: More Features from Cheap Operations

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

A Ghost Module is an image block for convolutional neural network that aims to generate more features by using fewer parameters. Specifically, an ordinary convolutional layer in deep neural networks is split into two parts. The first part involves ordinary convolutions but their total number is controlled. Given the intrinsic feature maps from the first part, a series of simple linear operations are applied for generating more feature maps.

Given the widely existing redundancy in intermediate feature maps calculated by mainstream CNNs, ghost modules aim to reduce them. In practice, given the input data X∈ℝ^(c×h×w), where c is the number of input channels and h and w are the height and width of the input data, respectively, the operation of an arbitrary convolutional layer for producing n feature maps can be formulated as

Y = X*f+b,

where * is the convolution operation, b is the bias term, Y∈ℝ^(h′×w′×n) is the output feature map with n channels, and f∈ℝ^(c×k×k ×n) is the convolution filters in this layer. In addition, h′ and w′ are the height and width of the output data, and k×k is the kernel size of convolution filters f, respectively. During this convolution procedure, the required number of FLOPs can be calculated as n·h′·w′·c·k·k, which is often as large as hundreds of thousands since the number of filters n and the channel number c are generally very large (e.g. 256 or 512).

Here, the number of parameters (in f and b) to be optimized is explicitly determined by the dimensions of input and output feature maps. The output feature maps of convolutional layers often contain much redundancy, and some of them could be similar with each other. We point out that it is unnecessary to generate these redundant feature maps one by one with large number of FLOPs and parameters. Suppose that the output feature maps are ghosts of a handful of intrinsic feature maps with some cheap transformations. These intrinsic feature maps are often of smaller size and produced by ordinary convolution filters. Specifically, m intrinsic feature maps Y′∈ℝ^(h′×w′×m) are generated using a primary convolution:

Y′ = X*f′,

where f′∈ℝ^(c×k×k ×m) is the utilized filters, m≤n and the bias term is omitted for simplicity. The hyper-parameters such as filter size, stride, padding, are the same as those in the ordinary convolution to keep the spatial size (ie h′ and w′) of the output feature maps consistent. To further obtain the desired n feature maps, we apply a series of cheap linear operations on each intrinsic feature in Y′ to generate s ghost features according to the following function:

yᵢⱼ = Φ_(i,j)(y′ᵢ), ∀ i = 1,...,m, j = 1,...,s,

where y′ᵢ is the i-th intrinsic feature map in Y′, Φ_(i,j) in the above function is the j-th (except the last one) linear operation for generating the j-th ghost feature map yᵢⱼ, that is to say, y′ᵢ can have one or more ghost feature maps {yᵢⱼ}ⱼ₌₁ˢ. The last Φ_(i,s) is the identity mapping for preserving the intrinsic feature maps. we can obtain n=m·s feature maps Y=[y₁₁,y₁₂,⋯,yₘₛ] as the output data of a Ghost module. Note that the linear operations Φ operate on each channel whose computational cost is much less than the ordinary convolution. In practice, there could be several different linear operations in a Ghost module, eg 3×3 and 5×5 linear kernels, which will be analyzed in the experiment part.

PaperSourceSee Code · huawei-noah/ghostnet

In Syntology code not checked for this name. The lookup matched a different name in Syntology's graph, 'GHOST' (resolution: token_match), and its answer is about that name only, so it does not say whether this method has code. About another paper: this answer is about arXiv:2502.03359, while the archive credits arXiv:1911.11907.

Code Syntology holds under other names, as the lookup lists them (code_nearby); not a complete list, and none of them is this method.

Origin paper of this answer, arXiv:2502.03359 (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: 30 · with a code link: 18 · where Syntology ran a sample: 5 (4 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 (5 of 30 tagged: 4 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 30 of 30, 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.

Tasks archive 2025-07-28

20 shown of 45 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
Image Classification8
Object Detection5
image-classification5
Defect Detection3
GPU3
Insulator Defect Detection3
Segmentation3
Semantic Segmentation3
Super-Resolution3
object-detection3
Image Segmentation2
Image Super-Resolution2
Model Compression2
Small Object Detection2
Acoustic Scene Classification1
Action Recognition1
Autonomous Driving1
CPU1
Data Augmentation1
Denoising1

Usage over time archive 2025-07-28

Papers per year tagged with Ghost Module: 2019 to 2025, peak 8 8 0 2019: 1 paper 2019 2020: 4 papers 2020 2021: 7 papers 2021 2022: 8 papers 2022 2023: 2 papers 2023 2024: 7 papers 2024 2025: 1 paper 2025
Papers per year the archive tags with this method, by the paper's archive date (30 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

Image Model Blocks

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