# Adaptively Connected Neural Networks

arXiv: https://arxiv.org/abs/1904.03579 · HTML: https://mlsanity.com/paper/1904.03579/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Apr 2019.

## Built on by (most cited first, 7 of 7)

- [Dynamic Neural Networks: A Survey](https://mlsanity.com/paper/2102.04906/) (2021)
- [SparseNeRF: Distilling Depth Ranking for Few-shot Novel View Synthesis](https://mlsanity.com/paper/2303.16196/) (2023)
- [AdaFocus V2: End-to-End Training of Spatial Dynamic Networks for Video Recognition](https://mlsanity.com/paper/2112.14238/) (2021)
- [TCGL: Temporal Contrastive Graph for Self-supervised Video Representation Learning](https://mlsanity.com/paper/2112.03587/) (2021)
- [Beyond Fixation: Dynamic Window Visual Transformer](https://mlsanity.com/paper/2203.12856/) (2022)
- [Uni-AdaFocus: Spatial-temporal Dynamic Computation for Video Recognition](https://mlsanity.com/paper/2412.11228/) (2024)
- [Onfocus Detection: Identifying Individual-Camera Eye Contact from Unconstrained Images](https://mlsanity.com/paper/2103.15307/) (2021)

## Abstract (arXiv)

This paper presents a novel adaptively connected neural network (ACNet) to improve the traditional convolutional neural networks (CNNs) in two aspects. First, ACNet employs a flexible way to switch global and local inference in processing the internal feature representations by adaptively determining the connection status among the feature nodes (e.g., pixels of the feature maps) footnoteIn a computer vision domain, a node refers to a pixel of a feature map, while in the graph domain, a node denotes a graph node.. We can show that existing CNNs, the classical multilayer perceptron (MLP), and the recently proposed non-local network (NLN) citenonlocalnn17 are all special cases of ACNet. Second, ACNet is also capable of handling non-Euclidean data. Extensive experimental analyses on a variety of benchmarks (i.e., ImageNet-1k classification, COCO 2017 detection and segmentation, CUHK03 person re-identification, CIFAR analysis, and Cora document categorization) demonstrate that ACNet cannot only achieve state-of-the-art performance but also overcome the limitation of the conventional MLP and CNN footnoteCorresponding author: Liang Lin (linliang@ieee.org). The code is available at urlhttps://github.com/wanggrun/Adaptively-Connected-Neural-Networks.