# Omni-Dimensional Dynamic Convolution

arXiv: https://arxiv.org/abs/2209.07947 · HTML: https://mlsanity.com/paper/2209.07947/

## Key facts

- Cited by 55 papers in the map (data through 14 Oct 2026).
- Builds on 41 of its 44 references that are in the map.
- First posted on arXiv 16 Sep 2022.

## Builds on (most cited first, 30 of 41)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [Aggregated Residual Transformations for Deep Neural Networks](https://mlsanity.com/paper/1611.05431/) (2016)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [Rich feature hierarchies for accurate object detection and semantic segmentation](https://mlsanity.com/paper/1311.2524/) (2013)
- [Searching for MobileNetV3](https://mlsanity.com/paper/1905.02244/) (2019)
- [Neural Architecture Search with Reinforcement Learning](https://mlsanity.com/paper/1611.01578/) (2016)
- [Spatial Transformer Networks](https://mlsanity.com/paper/1506.02025/) (2015)
- [MMDetection: Open MMLab Detection Toolbox and Benchmark](https://mlsanity.com/paper/1906.07155/) (2019)
- [ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices](https://mlsanity.com/paper/1707.01083/) (2017)
- [ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design](https://mlsanity.com/paper/1807.11164/) (2018)
- [Efficient Neural Architecture Search via Parameter Sharing](https://mlsanity.com/paper/1802.03268/) (2018)
- [ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks](https://mlsanity.com/paper/1910.03151/) (2019)
- [Selective Kernel Networks](https://mlsanity.com/paper/1903.06586/) (2019)
- [Residual Attention Network for Image Classification](https://mlsanity.com/paper/1704.06904/) (2017)
- [Dynamic Convolution: Attention over Convolution Kernels](https://mlsanity.com/paper/1912.03458/) (2019)
- [Dynamic Filter Networks](https://mlsanity.com/paper/1605.09673/) (2016)

## Built on by (most cited first, 30 of 55)

- [Merging Experts into One: Improving Computational Efficiency of Mixture of Experts](https://mlsanity.com/paper/2310.09832/) (2023)
- [Language Adaptive Weight Generation for Multi-task Visual Grounding](https://mlsanity.com/paper/2306.04652/) (2023)
- [Adaptive Dynamic Filtering Network for Image Denoising](https://mlsanity.com/paper/2211.12051/) (2022)
- [OverLoCK: An Overview-first-Look-Closely-next ConvNet with Context-Mixing Dynamic Kernels](https://mlsanity.com/paper/2502.20087/) (2025)
- [Frequency Dynamic Convolution for Dense Image Prediction](https://mlsanity.com/paper/2503.18783/) (2025)
- [LSNet: See Large, Focus Small](https://mlsanity.com/paper/2503.23135/) (2025)
- [Computational Optics for Mobile Terminals in Mass Production](https://mlsanity.com/paper/2305.05886/) (2023)
- [RefConv: Re-parameterized Refocusing Convolution for Powerful ConvNets](https://mlsanity.com/paper/2310.10563/) (2023)
- [FalconNet: Factorization for the Light-weight ConvNets](https://mlsanity.com/paper/2306.06365/) (2023)
- [Revealing the preference for correcting separated aberrations in joint optic-image design](https://mlsanity.com/paper/2309.04342/) (2023)
- [KernelWarehouse: Rethinking the Design of Dynamic Convolution](https://mlsanity.com/paper/2406.07879/) (2024)
- [MIPI 2023 Challenge on RGBW Remosaic: Methods and Results](https://mlsanity.com/paper/2304.10551/) (2023)
- [KernelWarehouse: Towards Parameter-Efficient Dynamic Convolution](https://mlsanity.com/paper/2308.08361/) (2023)
- [SRSNetwork: Siamese Reconstruction-Segmentation Networks based on Dynamic-Parameter Convolution](https://mlsanity.com/paper/2312.01741/) (2023)
- [RAW Image Reconstruction from RGB on Smartphones. NTIRE 2025 Challenge Report](https://mlsanity.com/paper/2506.01947/) (2025)
- [The Effectiveness of a Simplified Model Structure for Crowd Counting](https://mlsanity.com/paper/2404.07847/) (2024)
- [Metric Convolutions: A Unifying Theory to Adaptive Image Convolutions](https://mlsanity.com/paper/2406.05400/) (2024)
- [XS-VID: A Large-Scale Benchmark for Small Object Detection and Tracking in Videos](https://mlsanity.com/paper/2407.18137/) (2024)
- [RealOSR: Latent Guidance Boosts Diffusion-based Real-world Omnidirectional Image Super-Resolutions](https://mlsanity.com/paper/2412.09646/) (2024)
- [Dynamic Mobile-Former: Strengthening Dynamic Convolution with Attention and Residual Connection in Kernel Space](https://mlsanity.com/paper/2304.07254/) (2023)
- [Evolution: A Unified Formula for Feature Operators from a High-level Perspective](https://mlsanity.com/paper/2305.14409/) (2023)
- [D-YOLO a robust framework for object detection in adverse weather conditions](https://mlsanity.com/paper/2403.09233/) (2024)
- [Conv-INR: Convolutional Implicit Neural Representation for Multimodal Visual Signals](https://mlsanity.com/paper/2406.04249/) (2024)
- [Interactive Image-Based Aphid Counting in Yellow Water Traps under Stirring Actions](https://mlsanity.com/paper/2411.10357/) (2024)
- [HyPSAM: Hybrid Prompt-driven Segment Anything Model for RGB-Thermal Salient Object Detection](https://mlsanity.com/paper/2509.18738/) (2025)
- [DIGMN: Dynamic Intent Guided Meta Network for Differentiated User Engagement Forecasting in Online Professional Social Platforms](https://mlsanity.com/paper/2210.12402/) (2022)
- [DS-TDNN: Dual-stream Time-delay Neural Network with Global-aware Filter for Speaker Verification](https://mlsanity.com/paper/2303.11020/) (2023)
- [Single-Point Supervised High-Resolution Dynamic Network for Infrared Small Target Detection](https://mlsanity.com/paper/2408.01976/) (2024)
- [Heterogeneous Space Fusion and Dual-Dimension Attention: A New Paradigm for Speech Enhancement](https://mlsanity.com/paper/2408.06911/) (2024)
- [Multiscale Encoder and Omni-Dimensional Dynamic Convolution Enrichment in nnU-Net for Brain Tumor Segmentation](https://mlsanity.com/paper/2409.13229/) (2024)

## Abstract (arXiv)

Learning a single static convolutional kernel in each convolutional layer is the common training paradigm of modern Convolutional Neural Networks (CNNs). Instead, recent research in dynamic convolution shows that learning a linear combination of n convolutional kernels weighted with their input-dependent attentions can significantly improve the accuracy of light-weight CNNs, while maintaining efficient inference. However, we observe that existing works endow convolutional kernels with the dynamic property through one dimension (regarding the convolutional kernel number) of the kernel space, but the other three dimensions (regarding the spatial size, the input channel number and the output channel number for each convolutional kernel) are overlooked. Inspired by this, we present Omni-dimensional Dynamic Convolution (ODConv), a more generalized yet elegant dynamic convolution design, to advance this line of research. ODConv leverages a novel multi-dimensional attention mechanism with a parallel strategy to learn complementary attentions for convolutional kernels along all four dimensions of the kernel space at any convolutional layer. As a drop-in replacement of regular convolutions, ODConv can be plugged into many CNN architectures. Extensive experiments on the ImageNet and MS-COCO datasets show that ODConv brings solid accuracy boosts for various prevailing CNN backbones including both light-weight and large ones, e.g., 3.77%~5.71%|1.86%~3.72% absolute top-1 improvements to MobivleNetV2|ResNet family on the ImageNet dataset. Intriguingly, thanks to its improved feature learning ability, ODConv with even one single kernel can compete with or outperform existing dynamic convolution counterparts with multiple kernels, substantially reducing extra parameters. Furthermore, ODConv is also superior to other attention modules for modulating the output features or the convolutional weights.