Papers › Rethinking Depthwise Separable Convolutions: How Intra-Kernel Correlations Lead to...

Rethinking Depthwise Separable Convolutions: How Intra-Kernel Correlations Lead to Improved MobileNets

30 Mar 2020CVPR 2020 6arXiv:2003.13549archive 2025-07-28

Daniel Haase, Manuel Amthor

We introduce blueprint separable convolutions (BSConv) as highly efficient building blocks for CNNs. They are motivated by quantitative analyses of kernel properties from trained models, which show the dominance of correlations along the depth axis. Based on our findings, we formulate a theoretical foundation from which we derive efficient implementations using only standard layers. Moreover, our approach provides a thorough theoretical derivation, interpretation, and justification for the application of depthwise separable convolutions (DSCs) in general, which have become the basis of many modern network architectures. Ultimately, we reveal that DSC-based architectures such as MobileNets implicitly rely on cross-kernel correlations, while our BSConv formulation is based on intra-kernel correlations and thus allows for a more efficient separation of regular convolutions. Extensive experiments on large-scale and fine-grained classification datasets show that BSConvs clearly and consistently improve MobileNets and other DSC-based architectures without introducing any further complexity. For fine-grained datasets, we achieve an improvement of up to 13.7 percentage points. In addition, if used as drop-in replacement for standard architectures such as ResNets, BSConv variants also outperform their vanilla counterparts by up to 9.5 percentage points on ImageNet. Code and models are available under https://github.com/zeiss-microscopy/BSConv.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

zeiss-microscopy/BSConv officialmentioned in papermentioned on GitHubpytorchBSD-3-Clause-Clear report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Fine-Grained Image ClassificationImage Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Fine-Grained Image Classification Stanford Dogs EfficientNet-B0 (BSConv-S) Accuracy 61.2% #24 of 24 Archive leaderboard report
Image Classification CIFAR-100 MobileNetV3-large x1.0 (BSConv-U) Percentage correct 77.7 #141 of 211 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

1x1 ConvolutionDepthwise ConvolutionDepthwise Separable ConvolutionOrthogonal RegularizationPointwise Convolution

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