# Omni-Scale CNNs: a simple and effective kernel size configuration for time series classification

arXiv: https://arxiv.org/abs/2002.10061 · HTML: https://mlsanity.com/paper/2002.10061/

## Key facts

- Cited by 48 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Feb 2020.

## Built on by (most cited first, 30 of 48)

- [AnySat: One Earth Observation Model for Many Resolutions, Scales, and Modalities](https://mlsanity.com/paper/2412.14123/) (2024)
- [A Survey on Time-Series Pre-Trained Models](https://mlsanity.com/paper/2305.10716/) (2023)
- [TimeMAE: Self-Supervised Representations of Time Series with Decoupled Masked Autoencoders](https://mlsanity.com/paper/2303.00320/) (2023)
- [DIVERSIFY: A General Framework for Time Series Out-of-distribution Detection and Generalization](https://mlsanity.com/paper/2308.02282/) (2023)
- [Dynamic Sparse Network for Time Series Classification: Learning What to "see''](https://mlsanity.com/paper/2212.09840/) (2022)
- [FormerTime: Hierarchical Multi-Scale Representations for Multivariate Time Series Classification](https://mlsanity.com/paper/2302.09818/) (2023)
- [Label-efficient Time Series Representation Learning: A Review](https://mlsanity.com/paper/2302.06433/) (2023)
- [TodyNet: Temporal Dynamic Graph Neural Network for Multivariate Time Series Classification](https://mlsanity.com/paper/2304.05078/) (2023)
- [TimeMIL: Advancing Multivariate Time Series Classification via a Time-aware Multiple Instance Learning](https://mlsanity.com/paper/2405.03140/) (2024)
- [A Shapelet-based Framework for Unsupervised Multivariate Time Series Representation Learning](https://mlsanity.com/paper/2305.18888/) (2023)
- [Pixel-wise Agricultural Image Time Series Classification: Comparisons and a Deformable Prototype-based Approach](https://mlsanity.com/paper/2303.12533/) (2023)
- [PaAno: Patch-Based Representation Learning for Time-Series Anomaly Detection](https://mlsanity.com/paper/2602.01359/) (2026)
- [AimTS: Augmented Series and Image Contrastive Learning for Time Series Classification](https://mlsanity.com/paper/2504.09993/) (2025)
- [Fusing Structure from Motion and Simulation-Augmented Pose Regression from Optical Flow for Challenging Indoor Environments](https://mlsanity.com/paper/2304.07250/) (2023)
- [Evaluation of (Un-)Supervised Machine Learning Methods for GNSS Interference Classification with Real-World Data Discrepancies](https://mlsanity.com/paper/2503.23775/) (2025)
- [TimelyGPT: Extrapolatable Transformer Pre-training for Long-term Time-Series Forecasting in Healthcare](https://mlsanity.com/paper/2312.00817/) (2023)
- [TraffiDent: A Dataset for Understanding the Interplay Between Traffic Dynamics and Incidents](https://mlsanity.com/paper/2407.11477/) (2024)
- [Capturing Temporal Components for Time Series Classification](https://mlsanity.com/paper/2406.14456/) (2024)
- [Con4m: Context-aware Consistency Learning Framework for Segmented Time Series Classification](https://mlsanity.com/paper/2408.00041/) (2024)
- [DRFormer: Multi-Scale Transformer Utilizing Diverse Receptive Fields for Long Time-Series Forecasting](https://mlsanity.com/paper/2408.02279/) (2024)
- [A Large-Scale Multimodal Dataset and Benchmarks for Human Activity Scene Understanding and Reasoning](https://mlsanity.com/paper/2512.07136/) (2025)
- [A General Framework for Generative Self-supervised Learning in Non-invasive Estimation of Physiological Parameters Using Photoplethysmography](https://mlsanity.com/paper/2604.22780/) (2026)
- [Representation learning for a generalized, quantitative comparison of complex model outputs](https://mlsanity.com/paper/2208.06530/) (2022)
- [Enhancing Multivariate Time Series Classifiers through Self-Attention and Relative Positioning Infusion](https://mlsanity.com/paper/2302.06683/) (2023)
- [FedST: Secure Federated Shapelet Transformation for Time Series Classification](https://mlsanity.com/paper/2302.10631/) (2023)
- [Thwarting Cybersecurity Attacks with Explainable Concept Drift](https://mlsanity.com/paper/2403.13023/) (2024)
- [AdaFSNet: Time Series Classification Based on Convolutional Network with a Adaptive and Effective Kernel Size Configuration](https://mlsanity.com/paper/2404.18246/) (2024)
- [CAV-AD: A Robust Framework for Detection of Anomalous Data and Malicious Sensors in CAV Networks](https://mlsanity.com/paper/2407.05461/) (2024)
- [An End-to-End Model for Time Series Classification In the Presence of Missing Values](https://mlsanity.com/paper/2408.05849/) (2024)
- [Quasi Zigzag Persistence: A Topological Framework for Analyzing Time-Varying Data](https://mlsanity.com/paper/2502.16049/) (2025)

## Abstract (arXiv)

The Receptive Field (RF) size has been one of the most important factors for One Dimensional Convolutional Neural Networks (1D-CNNs) on time series classification tasks. Large efforts have been taken to choose the appropriate size because it has a huge influence on the performance and differs significantly for each dataset. In this paper, we propose an Omni-Scale block (OS-block) for 1D-CNNs, where the kernel sizes are decided by a simple and universal rule. Particularly, it is a set of kernel sizes that can efficiently cover the best RF size across different datasets via consisting of multiple prime numbers according to the length of the time series. The experiment result shows that models with the OS-block can achieve a similar performance as models with the searched optimal RF size and due to the strong optimal RF size capture ability, simple 1D-CNN models with OS-block achieves the state-of-the-art performance on four time series benchmarks, including both univariate and multivariate data from multiple domains. Comprehensive analysis and discussions shed light on why the OS-block can capture optimal RF sizes across different datasets. Code available [https://github.com/Wensi-Tang/OS-CNN]