Methods › Computer Vision › Image Model Blocks › Ghost Module
Ghost Module
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.
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.
- mixed ghost clipping · shares_name_tokens (shares name tokens with this name; a different method) · level V3 · retrieve with
get_reference_implementation(name='mixed ghost clipping')
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.
-
GRNN:Recurrent Neural Network based on Ghost Features for Video Super-Resolution 14 May 2025 · 0 repositories · arXiv:2505.10577
-
Cross-video Identity Correlating for Person Re-identification Pre-training 27 Sep 2024 · 1 repository · arXiv:2409.18569Syntology official: harvested, nothing ran · 0 ran · 1 unverified (of 1 harvested sample)
-
A Lightweight Insulator Defect Detection Model Based on Drone Images 26 Aug 2024 · 1 repository
-
IDD-YOLOv5: A Lightweight Insulator Defect Real-time Detection Algorithm 19 Aug 2024 · 1 repository
-
A lightweight YOLOv5-FFM model for occlusion pedestrian detection 13 Aug 2024 · 0 repositories · arXiv:2408.06633
-
LiteYOLO-ID: A Lightweight Object Detection Network for Insulator Defect Detection 24 Jun 2024 · 1 repository
-
Multimodal Emotion Recognition based on Facial Expressions, Speech, and EEG 11 Jun 2024 · 0 repositories
-
Ghost-Stereo: GhostNet-based Cost Volume Enhancement and Aggregation for Stereo Matching Networks 23 May 2024 · 0 repositories · arXiv:2405.14520
-
GRAN: Ghost Residual Attention Network for Single Image Super Resolution 28 Feb 2023 · 0 repositories · arXiv:2302.14557
-
Short-Term Memory Convolutions 8 Feb 2023 · 0 repositories · arXiv:2302.04331
-
GhostNetV2: Enhance Cheap Operation with Long-Range Attention 23 Nov 2022 · 12 repositories · arXiv:2211.12905Syntology 8 ran (of which 7 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified (of 10 harvested samples) · 10 pointer-only (licence)
-
RepGhost: A Hardware-Efficient Ghost Module via Re-parameterization 11 Nov 2022 · 3 repositories · arXiv:2211.06088Syntology official (archive's flag): 10 ran · 10 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 0 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 7 unverified (of 17 harvested samples)
-
Network Amplification With Efficient MACs Allocation 1 Jul 2022 · 2 repositories
-
YOLOv5s-GTB: light-weighted and improved YOLOv5s for bridge crack detection 3 Jun 2022 · 0 repositories · arXiv:2206.01498
-
MoCoViT: Mobile Convolutional Vision Transformer 25 May 2022 · 1 repository · arXiv:2205.12635
-
Efficient Convolutional Neural Networks on Raspberry Pi for Image Classification 2 Apr 2022 · 1 repository · arXiv:2204.00943Syntology official (archive's flag): 1 ran · 1 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; 1 where Syntology's instrument failed) · 0 unverified (of 1 harvested sample) · 1 pointer-only (licence)
-
ThreshNet: An Efficient DenseNet Using Threshold Mechanism to Reduce Connections 9 Jan 2022 · 1 repository · arXiv:2201.03013
-
GPU-Net: Lightweight U-Net with more diverse features 7 Jan 2022 · 1 repository · arXiv:2201.02656
-
Ghost-dil-NetVLAD: A Lightweight Neural Network for Visual Place Recognition 22 Dec 2021 · 0 repositories · arXiv:2112.11679
-
GhostShiftAddNet: More Features from Energy-Efficient Operations 20 Sep 2021 · 3 repositories · arXiv:2109.09495
-
GhostBERT: Generate More Features with Cheap Operations for BERT 1 Aug 2021 · 0 repositories
-
Greedy Network Enlarging 31 Jul 2021 · 1 repository · arXiv:2108.00177
-
AdaFuse: Adaptive Temporal Fusion Network for Efficient Action Recognition 10 Feb 2021 · 0 repositories · arXiv:2102.05775
-
Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet 28 Jan 2021 · 13 repositories · arXiv:2101.11986Syntology official (archive's flag): 4 ran · 21 ran (of which 16 constructed an object rather than computing a result; 21 with no instrument failure: 1 honoured, 0 violated, 20 with no contract checked; 0 where Syntology's instrument failed) · 5 unverified (of 26 harvested samples) · 8 pointer-only (licence)
-
GhostSR: Learning Ghost Features for Efficient Image Super-Resolution 21 Jan 2021 · 4 repositories · arXiv:2101.08525
-
A Multi-task Joint Framework for Real-time Person Search 11 Dec 2020 · 0 repositories · arXiv:2012.06418
-
Real-time Semantic Segmentation with Context Aggregation Network 2 Nov 2020 · 0 repositories · arXiv:2011.00993
-
Model Rubik's Cube: Twisting Resolution, Depth and Width for TinyNets 28 Oct 2020 · 9 repositories · arXiv:2010.14819Syntology official: harvested for another paper · 0 ran · 1 unverified (of 1 harvested sample) · 1 pointer-only (licence)
-
HS-ResNet: Hierarchical-Split Block on Convolutional Neural Network 15 Oct 2020 · 2 repositories · arXiv:2010.07621
-
GhostNet: More Features from Cheap Operations 27 Nov 2019 · 33 repositories · arXiv:1911.11907Syntology community repositories only · 19 ran (of which 0 constructed an object rather than computing a result; 16 with no instrument failure: 3 honoured, 0 violated, 13 with no contract checked; 3 where Syntology's instrument failed) · 4 unverified (of 23 harvested samples) · 5 pointer-only (licence)
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.
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