Dynamic Routing Between Capsules

A capsule is a group of neurons whose activity vector represents the instantiation parameters of a specific type of entity such as an object or an object part. We use the length of the activity vector to represent the probability that the entity exists and its orientation to represent the instantiation parameters. Active capsules at one level make predictions, via transformation matrices, for the instantiation parameters of higher-level capsules. When multiple predictions agree, a higher level capsule becomes active. We show that a discrimininatively trained, multi-layer capsule system achieves state-of-the-art performance on MNIST and is considerably better than a convolutional net at recognizing highly overlapping digits. To achieve these results we use an iterative routing-by-agreement mechanism: A lower-level capsule prefers to send its output to higher level capsules whose activity vectors have a big scalar product with the prediction coming from the lower-level capsule.

• Publication year

  2017

• Venue

  Neural Information Processing Systems

• Publication date

  2017-10-26

• Fields of study

  Computer Science

• Identifiers
  arXiv 1710.09829
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Tagger: Deep Unsupervised Perceptual Grouping

  2016cited by this paper
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems

  2016cited by this paper
• APAC: Augmented PAttern Classification with Neural Networks

  2015cited by this paper
• Spatial Transformer Networks

  2015cited by this paper
• Batch-normalized Maxout Network in Network

  2015cited by this paper
• Adam: A Method for Stochastic Optimization

  2014cited by this paper
• Multiple Object Recognition with Visual Attention

  2014cited by this paper
• Stochastic Pooling for Regularization of Deep Convolutional Neural Networks

  2013cited by this paper
• Regularization of Neural Networks using DropConnect

  2013cited by this paper
• Multi-digit Number Recognition from Street View Imagery using Deep Convolutional Neural Networks

  2013cited by this paper
• Multi-column deep neural networks for image classification

  2012influential reference
• Transforming Auto-Encoders

  2011influential reference
• Reading Digits in Natural Images with Unsupervised Feature Learning

  2011cited by this paper
• High-Performance Neural Networks for Visual Object Classification

  2011cited by this paper
• The mnist database of handwritten digits

  2005cited by this paper
• Crowding is unlike ordinary masking: distinguishing feature integration from detection.

  2004cited by this paper
• Learning methods for generic object recognition with invariance to pose and lighting

  2004cited by this paper
• Learning to Parse Images

  1999influential reference
• A neurobiological model of visual attention and invariant pattern recognition based on dynamic routing of information

  1993cited by this paper
• Shape Representation in Parallel Systems

  1981influential reference
• A Parallel Computation that Assigns Canonical Object-Based Frames of Reference

  1981cited by this paper

• Deep learning-based mechanism- and signal-driven framework for early prediction of milling surface roughness

  2026cites this paper
• GC-UNet: Efficient network for medical image segmentation

  2026cites this paper
• Zero-shot stance detection in practice: insights on training, prompting, and decoding with a capable lightweight LLM

  2026cites this paper
• Comparative Analysis of Deep Learning Techniques in Alzheimer’s Disease Diagnosis: Trends, Challenges, and Future Directions

  2026cites this paper
• Multilingual sentiment analysis in e-commerce customer reviews using GPT and deep learning-based weighted-ensemble model

  2026cites this paper
• Symmetry-Preserving Neural Indicators for Discontinuity Detection in High-Order Discontinuous Galerkin Solvers

  2026cites this paper
• Alzheimer’s disease prediction via an explainable CNN using genetic algorithm and SHAP values

  2026cites this paper
• Joint Identification of Encoders and Interleavers Using Deep Learning With Hardware Validation

  2026cites this paper
• DPCRec-dual preferences learning with capsule network for next POI recommendation

  2026cites this paper
• Providing projective and affine invariance for recognition by Multi-Angle-Scale Vision Transformer

  2026cites this paper
• Accelerated Multiple Wasserstein Gradient Flows for Multi-objective Distributional Optimization

  2026cites this paper
• Capsule Network for Prediction of Permeability Tensor from X-ray Images of Porous Media

  2026influential citation
• SynCaps-DDI: A hybrid neural model combining capsule networks and syntactic graph convolution for drug-drug interaction extraction

  2026cites this paper
• 基于Mamba的多尺度动态融合分类网络

  2026cites this paper
• BERT-OTA: Enhancing Hate Speech Detection With Ontology-Guided Transformer Attention

  2026cites this paper
• Sequential recommendation with long-short-term preference and side information

  2026cites this paper
• Hierarchical Shift Mixing - Beyond Dense Attention in Transformers

  2026cites this paper
• Non-target information also matters: InverseFormer tracker for single object tracking

  2026cites this paper
• On the Dynamics of Observation and Semantics

  2026cites this paper
• CapsNet-XAI: exploring the fusion of capsule networks and XAI for transparent leaf disease identification & classification

  2026cites this paper
• Multi-class Alzheimer’s disease (AD) classification using Swin Transformer wavelet and Gray Wolf Optimization (GWO)

  2026cites this paper
• Single- and multi-channel microseismic signal classification using dynamic-kernel attention-enhanced CNN-CapsNet

  2026cites this paper
• GRASFormer: A computational framework for gradient-regularized and entropy-stabilized multi-task transformer optimization

  2026cites this paper
• SEIS: Subspace-based Equivariance and Invariance Scores for Neural Representations

  2026cites this paper
• Spline-gated capsule networks with cross-dimensional transformations for nonlinear pose-aware recognition

  2026cites this paper
• CSAM: Capsule spatial attention mask network for visual question answering

  2026cites this paper
• NeiRNet: Neighborhood regularized network for retinal artery and vein classification in ultra-wide-field imaging

  2026cites this paper
• DFCCNet: Unified Dual-domain Fusion and Color-aware Residual Correction for Robust Single Image Dehazing

  2026cites this paper
• Inversion of heterogeneous hydrogeological parameters using convolutional adversarial autoencoder and deep capsule encoder-decoder network with ILUES

  2026influential citation
• Mathematical Frameworks in Image Captioning: A Comprehensive Survey and Real-Time Processing Analysis

  2026cites this paper
• Hybrid Capsule Networks for early Parkinson's disease detection: Reclassifying SWEDD cases through SPECT image analysis

  2026cites this paper
• Research progress on emotion recognition based on electroencephalogram signals

  2026cites this paper
• One-Shot Identification with Different Neural Network Approaches

  2026influential citation
• Adaptive group-weighted convolutions: Enhancing equivariant networks via learnable symmetry importance maps

  2026cites this paper
• Transformer based Multi-task Fusion Network for Food Spoilage Detection and Shelf life Forecasting

  2026cites this paper
• Latent Object Permanence: Topological Phase Transitions, Free-Energy Principles, and Renormalization Group Flows in Deep Transformer Manifolds

  2026cites this paper
• BiERU-CapsNet: Bidirectional Emotional Recurrent Unit with Capsule Network, a Novel Approach for Textual Sentiment Analysis

  2026cites this paper
• Pump stroke event-driven cross-cylinder transfer learning for fluid-end fault diagnosis in multi-cylinder drilling pumps

  2026influential citation
• Interpreting multi-agent reinforcement learning decisions via key feature activation

  2026cites this paper
• Efficient sarcasm detection in social media using hybrid CapsNet-LSTM fusion and feature optimization

  2026influential citation
• Lightweight attentional convolutional capsule network for facial expression recognition

  2026influential citation
• DeepFake detection in images and videos: a survey on models, datasets, and evaluation metrics

  2026cites this paper
• Sustainable Maritime Applications with Lightweight Classifier Using Modified MobileNet

  2026cites this paper
• Effectiveness of Capsule Networks in Detecting Deepfakes Instead of Traditional CNNs

  2026cites this paper
• Quality-adaptive late fusion for capsule-based audio-visual speech recognition

  2026cites this paper
• Hierarchical capsule networks with domain-specific pre-training for robust audio-visual speech recognition

  2026cites this paper
• A Study on Real-time Object Detection using Deep Learning

  2026cites this paper
• A comprehensive review of convolutional neural networks: foundations, enhancements and applications

  2026cites this paper
• Emo-STCapsnet: A spatio-temporal modeling approach with enhanced CapsNet for speech emotion recognition

  2026cites this paper
• Estimation of the height of monocular satellite remote sensing images based on semantic information guidance

  2025cites this paper
• Comparative Study of Unimodal and Multimodal Systems Based on MNN

  2025cites this paper
• PDR: a language model for predicting protein-DNA binding sites

  2025cites this paper
• Improving Medical Waste Classification with Hybrid Capsule Networks

  2025cites this paper
• Emotion recognition and interaction of smart education environment screen based on deep learning networks

  2025cites this paper
• Spectral State Space Model for Rotation-Invariant Visual Representation Learning

  2025cites this paper
• Plagiarism detection for panoramic images

  2025cites this paper
• i-WiViG: Interpretable Window Vision GNN

  2025cites this paper
• A Dual-Stream Deep Learning Architecture With Adaptive Random Vector Functional Link for Multi-Center Ischemic Stroke Classification

  2025cites this paper
• CapT: A Hierarchical Capsule Representation Learning Approach for Class Continual Learning

  2025influential citation
• Crop Disease Detection using Hybrid Model of CNN and Capsule Networks

  2025cites this paper
• A Novel Acoustic-Based Framework for Compound Fault Diagnosis in Rotating Machinery With Limited Samples

  2025cites this paper
• Evaluating Backbone Modifications on Capsule Networks for Low-Resolution Image Classification

  2025influential citation
• Resource-Efficient and Noise-Robust Modality Fusion for Audio-Visual Speech Recognition

  2025cites this paper
• Boosting Semi-Supervised Video Action Detection with Temporal Context

  2025cites this paper
• A visual SLAM loop closure detection method based on lightweight siamese capsule network

  2025cites this paper
• Spatial Orientation Relation Recognition for Water Surface Targets

  2025cites this paper
• Train on classical, deploy on quantum: scaling generative quantum machine learning to a thousand qubits

  2025cites this paper
• Flexible Framework for Lung and Colon Cancer Automated Analysis Across Multiple Diagnosis Scenarios

  2025cites this paper
• Enhancing deep neural networks through complex-valued representations and Kuramoto synchronization dynamics

  2025cites this paper
• Enhancing Road Accident Classification with Capsule Recurrent Neural Network and Improved Reptile Search Algorithm

  2025cites this paper
• A Comparative Analysis of Compression and Transfer Learning Techniques in DeepFake Detection Models

  2025cites this paper
• High-quality control of receiver functions using a capsule neural network

  2025cites this paper
• Micro-expression detection in ASD movies: a YOLOv8-SMART approach

  2025cites this paper
• Leveraging Deep Q-Network Agents with Dynamic Routing Mechanisms in Convolutional Neural Networks for Enhanced and Reliable Classification of Alzheimer's Disease from MRI Scans

  2025cites this paper
• Semantic Gaussian Mixture Variational Autoencoder for Sequential Recommendation

  2025cites this paper
• Benchmarking Android Malware Detection: Rethinking the Role of Traditional and Deep Learning Models

  2025cites this paper
• AlexCapsNet: An Integrated Architecture for Image Classification With Background Noise

  2025influential citation
• Leveraging Capsule Networks for Robust Brain Tumor Classification and Detection in MRI Scans

  2025cites this paper
• TCrossE: Cross-space interaction of bicomplex and quaternion embeddings for temporal knowledge graph completion

  2025cites this paper
• NdLinear: Preserving Multi-Dimensional Structure for Parameter-Efficient Neural Networks

  2025cites this paper
• Capsule feature selector for software defect prediction

  2025influential citation
• Music genre classification using deep neural networks and data augmentation

  2025cites this paper
• CapsF: Capsule Fusion for Extracting psychiatric stressors for suicide from Twitter

  2025cites this paper
• Online deep learning’s role in conquering the challenges of streaming data: a survey

  2025cites this paper
• BandFocusNet: A Lightweight Model for Motor Imagery Classification of a Supernumerary Thumb in Virtual Reality

  2025cites this paper
• Efficient Self-Supervised Neural Architecture Search

  2025cites this paper
• Generative Artificial Intelligence and the Evolving Challenge of Deepfake Detection: A Systematic Analysis

  2025cites this paper
• CoRPA: Adversarial Image Generation for Chest X-rays Using Concept Vector Perturbations and Generative Models

  2025cites this paper
• A Multilevel Attitude-Aware Denoising Network for Bearing Fault Diagnosis

  2025influential citation
• Contextual and uncertainty-aware approach for multi-person pose estimation

  2025cites this paper
• Capsule network with using shifted windows for 3D human pose estimation

  2025cites this paper
• Adaptive user multi-level and multi-interest preferences for sequential recommendation

  2025cites this paper
• JSSAA-Net: A Joint Sparse Sampling and Attention Augmented ISAR Image Classification Network

  2025cites this paper
• Enhancing Malaria Detection Through Deep Learning: A Comparative Study of Convolutional Neural Networks

  2025cites this paper
• Noise filtering approach to improve handwritten digit recognition using customized CNN for Cerebral Palsy individuals

  2025cites this paper
• Intelligent chatter detection in high-speed milling using successive variational mode decomposition and a multi-channel feature fusion network

  2025cites this paper
• Generative artificial intelligence: a historical perspective

  2025cites this paper
• Traveling Waves Integrate Spatial Information Through Time

  2025cites this paper
• Data reduction for black-box adversarial attacks against deep neural networks based on side-channel attacks

  2025cites this paper
• GazeCapsNet: A Lightweight Gaze Estimation Framework

  2025cites this paper