$ShiftwiseConv:$ Small Convolutional Kernel with Large Kernel Effect Article Swipe

MAKE IMPACT

Dachong Li , Li Li , Zhuangzhuang Chen , Jianqiang Li · YOU? · · 2024 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2401.12736

Large kernels make standard convolutional neural networks (CNNs) great again over transformer architectures in various vision tasks. Nonetheless, recent studies meticulously designed around increasing kernel size have shown diminishing returns or stagnation in performance. Thus, the hidden factors of large kernel convolution that affect model performance remain unexplored. In this paper, we reveal that the key hidden factors of large kernels can be summarized as two separate components: extracting features at a certain granularity and fusing features by multiple pathways. To this end, we leverage the multi-path long-distance sparse dependency relationship to enhance feature utilization via the proposed Shiftwise (SW) convolution operator with a pure CNN architecture. In a wide range of vision tasks such as classification, segmentation, and detection, SW surpasses state-of-the-art transformers and CNN architectures, including SLaK and UniRepLKNet. More importantly, our experiments demonstrate that $3 \times 3$ convolutions can replace large convolutions in existing large kernel CNNs to achieve comparable effects, which may inspire follow-up works. Code and all the models at https://github.com/lidc54/shift-wiseConv.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2401.12736

PDF

https://arxiv.org/pdf/2401.12736

OA Status

green

Cited By

8

Related Works

10

OpenAlex ID

https://openalex.org/W4391212690

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4391212690

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2401.12736

Digital Object Identifier

Title

$ShiftwiseConv:$ Small Convolutional Kernel with Large Kernel Effect

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2024

Year of publication

Publication date

2024-01-23

Full publication date if available

Authors

Dachong Li, Li Li, Zhuangzhuang Chen, Jianqiang Li

List of authors in order

Landing page

https://arxiv.org/abs/2401.12736

Publisher landing page

PDF URL

https://arxiv.org/pdf/2401.12736

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2401.12736

Direct OA link when available

Concepts

Convolutional neural network, Computer science, Kernel (algebra), Convolutional code, Convolution (computer science), Operator (biology), Artificial intelligence, Pattern recognition (psychology), Algorithm, Artificial neural network, Mathematics, Repressor, Transcription factor, Chemistry, Combinatorics, Gene, Decoding methods, Biochemistry

Top concepts (fields/topics) attached by OpenAlex

Cited by

8

Total citation count in OpenAlex

Citations by year (recent)

2025: 3, 2024: 5

Per-year citation counts (last 5 years)

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload