# Global Aggregation then Local Distribution in Fully Convolutional Networks

arXiv: https://arxiv.org/abs/1909.07229 · HTML: https://mlsanity.com/paper/1909.07229/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Sep 2019.

## Built on by (most cited first, 22 of 22)

- [MST++: Multi-stage Spectral-wise Transformer for Efficient Spectral Reconstruction](https://mlsanity.com/paper/2204.07908/) (2022)
- [Coarse-to-Fine Sparse Transformer for Hyperspectral Image Reconstruction](https://mlsanity.com/paper/2203.04845/) (2022)
- [Enhanced Boundary Learning for Glass-like Object Segmentation](https://mlsanity.com/paper/2103.15734/) (2021)
- [Flow-Guided Sparse Transformer for Video Deblurring](https://mlsanity.com/paper/2201.01893/) (2022)
- [Loss Functions in the Era of Semantic Segmentation: A Survey and Outlook](https://mlsanity.com/paper/2312.05391/) (2023)
- [PointFlow: Flowing Semantics Through Points for Aerial Image Segmentation](https://mlsanity.com/paper/2103.06564/) (2021)
- [DuAT: Dual-Aggregation Transformer Network for Medical Image Segmentation](https://mlsanity.com/paper/2212.11677/) (2022)
- [SPIN Road Mapper: Extracting Roads from Aerial Images via Spatial and Interaction Space Graph Reasoning for Autonomous Driving](https://mlsanity.com/paper/2109.07701/) (2021)
- [Fashionformer: A simple, Effective and Unified Baseline for Human Fashion Segmentation and Recognition](https://mlsanity.com/paper/2204.04654/) (2022)
- [Fully Attentional Network for Semantic Segmentation](https://mlsanity.com/paper/2112.04108/) (2021)
- [Global Aggregation then Local Distribution for Scene Parsing](https://mlsanity.com/paper/2107.13154/) (2021)
- [Vision Transformers: From Semantic Segmentation to Dense Prediction](https://mlsanity.com/paper/2207.09339/) (2022)
- [BoundarySqueeze: Image Segmentation as Boundary Squeezing](https://mlsanity.com/paper/2105.11668/) (2021)
- [Transferable Attack for Semantic Segmentation](https://mlsanity.com/paper/2307.16572/) (2023)
- [Semantic Segmentation via Pixel-to-Center Similarity Calculation](https://mlsanity.com/paper/2301.04870/) (2023)
- [The Ikshana Hypothesis of Human Scene Understanding](https://mlsanity.com/paper/2101.10837/) (2021)
- [On the Effect of Image Resolution on Semantic Segmentation](https://mlsanity.com/paper/2402.05398/) (2024)
- [Few-Shot Fruit Segmentation via Transfer Learning](https://mlsanity.com/paper/2405.02556/) (2024)
- [Global and Local Features through Gaussian Mixture Models on Image Semantic Segmentation](https://mlsanity.com/paper/2207.09162/) (2022)
- [Dynamic Graph Message Passing Networks for Visual Recognition](https://mlsanity.com/paper/2209.09760/) (2022)
- [Efficient Context Integration through Factorized Pyramidal Learning for Ultra-Lightweight Semantic Segmentation](https://mlsanity.com/paper/2302.11785/) (2023)
- [AttenScribble: Attentive Similarity Learning for Scribble-Supervised Medical Image Segmentation](https://mlsanity.com/paper/2312.06614/) (2023)

## Abstract (arXiv)

It has been widely proven that modelling long-range dependencies in fully convolutional networks (FCNs) via global aggregation modules is critical for complex scene understanding tasks such as semantic segmentation and object detection. However, global aggregation is often dominated by features of large patterns and tends to oversmooth regions that contain small patterns (e.g., boundaries and small objects). To resolve this problem, we propose to first use Global Aggregation and then Local Distribution, which is called GALD, where long-range dependencies are more confidently used inside large pattern regions and vice versa. The size of each pattern at each position is estimated in the network as a per-channel mask map. GALD is end-to-end trainable and can be easily plugged into existing FCNs with various global aggregation modules for a wide range of vision tasks, and consistently improves the performance of state-of-the-art object detection and instance segmentation approaches. In particular, GALD used in semantic segmentation achieves new state-of-the-art performance on Cityscapes test set with mIoU 83.3%. Code is available at: urlhttps://github.com/lxtGH/GALD-Net