# Gated Channel Transformation for Visual Recognition

arXiv: https://arxiv.org/abs/1909.11519 · HTML: https://mlsanity.com/paper/1909.11519/

## Key facts

- Cited by 43 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Sep 2019.

## Built on by (most cited first, 30 of 43)

- [ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders](https://mlsanity.com/paper/2301.00808/) (2023)
- [Attention Mechanisms in Computer Vision: A Survey](https://mlsanity.com/paper/2111.07624/) (2021)
- [Salient Object Detection via Integrity Learning](https://mlsanity.com/paper/2101.07663/) (2021)
- [MIGC: Multi-Instance Generation Controller for Text-to-Image Synthesis](https://mlsanity.com/paper/2402.05408/) (2024)
- [Dynamic Slimmable Network](https://mlsanity.com/paper/2103.13258/) (2021)
- [NTIRE 2021 Challenge on High Dynamic Range Imaging: Dataset, Methods and Results](https://mlsanity.com/paper/2106.01439/) (2021)
- [Reliable Propagation-Correction Modulation for Video Object Segmentation](https://mlsanity.com/paper/2112.02853/) (2021)
- [Deep Digging into the Generalization of Self-Supervised Monocular Depth Estimation](https://mlsanity.com/paper/2205.11083/) (2022)
- [Augmenting Deep Classifiers with Polynomial Neural Networks](https://mlsanity.com/paper/2104.07916/) (2021)
- [Scalable Video Object Segmentation with Identification Mechanism](https://mlsanity.com/paper/2203.11442/) (2022)
- [DSC-PoseNet: Learning 6DoF Object Pose Estimation via Dual-scale Consistency](https://mlsanity.com/paper/2104.03658/) (2021)
- [FactorizePhys: Matrix Factorization for Multidimensional Attention in Remote Physiological Sensing](https://mlsanity.com/paper/2411.01542/) (2024)
- [Object Segmentation by Mining Cross-Modal Semantics](https://mlsanity.com/paper/2305.10469/) (2023)
- [Towards Robust Video Object Segmentation with Adaptive Object Calibration](https://mlsanity.com/paper/2207.00887/) (2022)
- [EEG motor imagery decoding: A framework for comparative analysis with channel attention mechanisms](https://mlsanity.com/paper/2310.11198/) (2023)
- [SCSA: Exploring the Synergistic Effects Between Spatial and Channel Attention](https://mlsanity.com/paper/2407.05128/) (2024)
- [Dynamic Slimmable Denoising Network](https://mlsanity.com/paper/2110.08940/) (2021)
- [Co-attention Propagation Network for Zero-Shot Video Object Segmentation](https://mlsanity.com/paper/2304.03910/) (2023)
- [Explore Internal and External Similarity for Single Image Deraining with Graph Neural Networks](https://mlsanity.com/paper/2406.00721/) (2024)
- [DS-Net++: Dynamic Weight Slicing for Efficient Inference in CNNs and Transformers](https://mlsanity.com/paper/2109.10060/) (2021)
- [Transformers Meet Visual Learning Understanding: A Comprehensive Review](https://mlsanity.com/paper/2203.12944/) (2022)
- [Booster-SHOT: Boosting Stacked Homography Transformations for Multiview Pedestrian Detection with Attention](https://mlsanity.com/paper/2208.09211/) (2022)
- [A Study on the Generality of Neural Network Structures for Monocular Depth Estimation](https://mlsanity.com/paper/2301.03169/) (2023)
- [Dynamic Semantic Compression for CNN Inference in Multi-access Edge Computing: A Graph Reinforcement Learning-based Autoencoder](https://mlsanity.com/paper/2401.12167/) (2024)
- [Uncertainty Guided Refinement for Fine-Grained Salient Object Detection](https://mlsanity.com/paper/2504.09666/) (2025)
- [AttZoom: Attention Zoom for Better Visual Features](https://mlsanity.com/paper/2508.03625/) (2025)
- [Ultra-high Resolution Image Segmentation via Locality-aware Context Fusion and Alternating Local Enhancement](https://mlsanity.com/paper/2109.02580/) (2021)
- [Mix-Pooling Strategy for Attention Mechanism](https://mlsanity.com/paper/2208.10322/) (2022)
- [Attention-based Feature Compression for CNN Inference Offloading in Edge Computing](https://mlsanity.com/paper/2211.13745/) (2022)
- [Hierarchical Graph Pattern Understanding for Zero-Shot VOS](https://mlsanity.com/paper/2312.09525/) (2023)

## Abstract (arXiv)

In this work, we propose a generally applicable transformation unit for visual recognition with deep convolutional neural networks. This transformation explicitly models channel relationships with explainable control variables. These variables determine the neuron behaviors of competition or cooperation, and they are jointly optimized with the convolutional weight towards more accurate recognition. In Squeeze-and-Excitation (SE) Networks, the channel relationships are implicitly learned by fully connected layers, and the SE block is integrated at the block-level. We instead introduce a channel normalization layer to reduce the number of parameters and computational complexity. This lightweight layer incorporates a simple l2 normalization, enabling our transformation unit applicable to operator-level without much increase of additional parameters. Extensive experiments demonstrate the effectiveness of our unit with clear margins on many vision tasks, i.e., image classification on ImageNet, object detection and instance segmentation on COCO, video classification on Kinetics.