Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps

This paper addresses the visualisation of image classification models, learnt using deep Convolutional Networks (ConvNets). We consider two visualisation techniques, based on computing the gradient of the class score with respect to the input image. The first one generates an image, which maximises the class score [Erhan et al., 2009], thus visualising the notion of the class, captured by a ConvNet. The second technique computes a class saliency map, specific to a given image and class. We show that such maps can be employed for weakly supervised object segmentation using classification ConvNets. Finally, we establish the connection between the gradient-based ConvNet visualisation methods and deconvolutional networks [Zeiler et al., 2013].

• Publication year

  2013

• Venue

  International Conference on Learning Representations

• Publication date

  2013-12-20

• Fields of study

  Computer Science

• Identifiers
  arXiv 1312.6034
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Visualizing and Understanding Convolutional Networks

  2013influential reference
• Multi-column deep neural networks for image classification

  2012cited by this paper
• ImageNet classification with deep convolutional neural networks

  2012influential reference
• Building high-level features using large scale unsupervised learning

  2011cited by this paper
• Improving the Fisher Kernel for Large-Scale Image Classification

  2010cited by this paper
• How to Explain Individual Classification Decisions

  2009cited by this paper
• Visualizing Higher-Layer Features of a Deep Network

  2009influential reference
• A discriminatively trained, multiscale, deformable part model

  2008cited by this paper
• A Fast Learning Algorithm for Deep Belief Nets

  2006cited by this paper
• Interactive graph cuts for optimal boundary & region segmentation of objects in N-D images

  2001cited by this paper
• Gradient-based learning applied to document recognition

  1998cited by this paper

• Counterfactual Explanations on Robust Perceptual Geodesics

  2026cites this paper
• Quantifying Explanation Quality in Graph Neural Networks using Out-of-Distribution Generalization

  2026cites this paper
• A deep learning framework for species-level identification of root-knot nematodes using perineal pattern images

  2026cites this paper
• Atomic Depth Estimation From Noisy Electron Microscopy Data Via Deep Learning

  2026cites this paper
• Investigating the Utility of Explainable Artificial Intelligence for Neuroimaging‐Based Dementia Diagnosis and Prognosis

  2026cites this paper
• Evaluating Object-Centric Models beyond Object Discovery

  2026cites this paper
• Explore the Ideology of Deep Learning in ENSO Forecasts

  2026cites this paper
• Uncertainty-aware genomic deep learning with knowledge distillation

  2026influential citation
• Insights on the Working Principles of a CNN for Forest Height Regression From Single-Pass InSAR Data

  2026cites this paper
• UniCog: Uncovering Cognitive Abilities of LLMs through Latent Mind Space Analysis

  2026cites this paper
• Explainable AI-Driven Quality and Condition Monitoring in Smart Manufacturing

  2026cites this paper
• Context Dependence and Reliability in Autoregressive Language Models

  2026cites this paper
• DAVE: Distribution-aware Attribution via ViT Gradient Decomposition

  2026cites this paper
• Exploring SAIG Methods for an Objective Evaluation of XAI

  2026cites this paper
• Time Series-Based Explainable Model for Lithium-Ion Battery State of Health Prediction

  2026cites this paper
• FaithSCAN: Model-Driven Single-Pass Hallucination Detection for Faithful Visual Question Answering

  2026cites this paper
• Inference Attacks Against Graph Generative Diffusion Models

  2026cites this paper
• RoadMark-cGAN: Generative Conditional Learning to Directly Map Road Marking Lines from Aerial Orthophotos via Image-to-Image Translation

  2026cites this paper
• Hierarchical Pooling and Explainability in Graph Neural Networks for Tumor and Tissue-of-Origin Classification Using RNA-seq Data

  2026cites this paper
• EvoMorph: Counterfactual Explanations for Continuous Time-Series Extrinsic Regression Applied to Photoplethysmography

  2026cites this paper
• Toward Faithful Explanations in Acoustic Anomaly Detection

  2026cites this paper
• Path-sampled integrated gradients

  2026cites this paper
• Assembling the Mind's Mosaic: Towards EEG Semantic Intent Decoding

  2026cites this paper
• A novel explainable deep-learning approach for network analysis of epistatic interactions

  2026cites this paper
• OOVDet: Low-Density Prior Learning for Zero-Shot Out-of-Vocabulary Object Detection

  2026cites this paper
• Unraveling the Hidden Dynamical Structure in Recurrent Neural Policies

  2026cites this paper
• A hyperspectral co-design framework guided by occlusion sensitivity for early mould detection in bamboo.

  2026cites this paper
• Do Prompts Guarantee Safety? Mitigating Toxicity from LLM Generations through Subspace Intervention

  2026cites this paper
• RTS: Towards Underground Root Tuber Sensing via RF Sensor Networks

  2026cites this paper
• FreqLens: Interpretable Frequency Attribution for Time Series Forecasting

  2026cites this paper
• Multisite T1-weighted MRI classification of Alzheimer's disease using 3D-CNN-HSCAM architecture with contrastive domain adaptation

  2026cites this paper
• EGMA-Net: Edge-guided multi-branch attention fusion network for gestational age prediction from fetal MRI

  2026cites this paper
• Point-ITR: Task-Oriented Importance Sampling for Large-Scale 3D Point Clouds in Manufacturing

  2026cites this paper
• Bridging Explainability and Accuracy in CLIP for Object Recognition via Joint Probability Over Rationales and Categories

  2026cites this paper
• Detecting Hallucinations in Retrieval-Augmented Generation via Semantic-level Internal Reasoning Graph

  2026cites this paper
• DeepLeak: Privacy Enhancing Hardening of Model Explanations Against Membership Leakage

  2026cites this paper
• Parallelizing Node-Level Explainability in Graph Neural Networks

  2026influential citation
• ENSO‐CausalNet: Integrating Causal Inference Into Deep Learning for Robust ENSO Prediction

  2026cites this paper
• Development of a Measurement Procedure for Emotional States Detection Based on Single-Channel Ear-EEG: A Proof-of-Concept Study

  2026cites this paper
• Hidden Monotonicity: Explaining Deep Neural Networks via their DC Decomposition

  2026cites this paper
• Image Style Transfer–Based High- and Low-Frequency Component Synthesis Neural Architecture for Low-Dose Computed Tomography Denoising

  2026cites this paper
• Load torque estimation in a three-phase induction motor through a data-driven sensor fusion approach

  2026cites this paper
• Feature-Aware Test Generation for Deep Learning Models

  2026cites this paper
• MATEX: Multi-scale Attention and Text-guided Explainability of Medical Vision-Language Models

  2026cites this paper
• Dialectal substitution as an adversarial approach for evaluating Arabic NLP robustness

  2026cites this paper
• IG-3D: Integrated-Gradients 3D Optimization for Private Transformer Inference

  2026cites this paper
• Interpretability of the Intent Detection Problem: A New Approach

  2026cites this paper
• Memory Retrieval in Transformers: Insights from The Encoding Specificity Principle

  2026cites this paper
• From Basins to safe sets: a machine learning perspective on chaotic dynamics

  2026cites this paper
• How ‘Neural’ is a Neural Foundation Model?

  2026cites this paper
• Large Language Model Agents Are Not Always Faithful Self-Evolvers

  2026cites this paper
• Contrastive Learning of Extragalactic Stellar Streams: Sculpting a Latent Space of Representations with DES DR2 Photometry

  2026influential citation
• Metaheuristic Hyperparameter Optimization and Explainable Deep Learning for Baggage Threat Detection

  2026cites this paper
• Toward Interpretable and Generalizable AI in Regulatory Genomics

  2026cites this paper
• Auditing Sybil: Explaining Deep Lung Cancer Risk Prediction Through Generative Interventional Attributions

  2026cites this paper
• Shortcut Features as Top Eigenfunctions of NTK: A Linear Neural Network Case and More

  2026cites this paper
• BAED: A new paradigm for few-shot graph learning with explanation in the loop.

  2026cites this paper
• A dual-strategy multi-objective evolutionary algorithm with domain-informed mutation and secondary weighted selection for counterfactual explanations in knowledge tracing

  2026cites this paper
• From Features to Actions: Explainability in Traditional and Agentic AI Systems

  2026cites this paper
• Explainable Artificial Intelligence for Deepfake Detection: Pipeline, Open Source and Comparisons

  2026cites this paper
• Is Meta-Path Attention an Explanation? Evidence of Alignment and Decoupling in Heterogeneous GNNs

  2026cites this paper
• Where Not to Learn: Prior-Aligned Training with Subset-based Attribution Constraints for Reliable Decision-Making

  2026cites this paper
• Visual Question Explainable Reasoning on Hypothesis Agent Interaction with Scene

  2026cites this paper
• KAN-Based Interpretable Radio Map Prediction Framework With Symbolic Data Fusion

  2026cites this paper
• HEEDFUL: Leveraging Sequential Transfer Learning for Robust WiFi Device Fingerprinting Amid Hardware Warm-Up Effects

  2026cites this paper
• Predicting pyrazinamide resistance in Mycobacterium tuberculosis using a graph convolutional network

  2026cites this paper
• VCHAR: A Variance-Driven Framework for Complex Human Activity Recognition With Vague Label

  2026cites this paper
• SynthraXCoreNet: An Interpretable, Well-Calibrated Six-CNN Ensemble for Dermoscopic Skin-Lesion Classification

  2026influential citation
• Interpret When Possible: A Tree-Based Hybrid Framework for Interpretable Classification

  2026cites this paper
• PhiNet: Speaker Verification With Phonetic Interpretability

  2026cites this paper
• AI-Guided Inference of Morphodynamic Attractor-like States in Glioblastoma

  2026cites this paper
• Applicability study of AI attribution methods for ophthalmic image classification

  2026cites this paper
• Grad-ELLM: Gradient-based Explanations for Decoder-only LLMs

  2026influential citation
• In Search of Grandmother Cells: Tracing Interpretable Neurons in Tabular Representations

  2026cites this paper
• Deep behavioral phenotyping of pathogen infected mosquitoes reveals species-specific behavior changes enhancing transmission

  2026cites this paper
• Aligned explanations in neural networks

  2026cites this paper
• Learning to Reason: Temporal Saliency Distillation for Interpretable Knowledge Transfer

  2026cites this paper
• Clustering of Temporal and Visual Data: Recent Advancements

  2026cites this paper
• Explainable AI: Learning from the Learners

  2026cites this paper
• Optimising Image Feature Extraction and Selection: A Comprehensive Review With Spark Case Studies

  2026cites this paper
• Two Pathways to Truthfulness: On the Intrinsic Encoding of LLM Hallucinations

  2026cites this paper
• Outcome-Grounded Advantage Reshaping for Fine-Grained Credit Assignment in Mathematical Reasoning

  2026cites this paper
• An Evaluation of XAI Methods applied to Information Disorder Detection Models

  2026cites this paper
• PosIR: Position-Aware Heterogeneous Information Retrieval Benchmark

  2026cites this paper
• Hot-Start from Pixels: Low-Resolution Visual Tokens for Chinese Language Modeling

  2026cites this paper
• Toward Understanding Unlearning Difficulty: A Mechanistic Perspective and Circuit-Guided Difficulty Metric

  2026cites this paper
• AI-powered Biomedical Imaging: Recent Achievements and Challenges

  2026cites this paper
• Deep learning model design for ECG-based prediction of sex and ventricular dysfunction: A technical exploration of CNN and transformer architectures

  2026cites this paper
• Concepts from Representations: Post-hoc Concept Bottleneck Models via Sparse Decomposition of Visual Representations

  2026cites this paper
• Towards Visually Explaining Statistical Tests with Applications in Biomedical Imaging

  2026cites this paper
• Tracing 3D Anatomy in 2D Strokes: A Multi-Stage Projection Driven Approach to Cervical Spine Fracture Identification

  2026cites this paper
• DDSA: Dual-Domain Strategic Attack for Spatial-Temporal Efficiency in Adversarial Robustness Testing

  2026cites this paper
• Bridging the Black Box: A Survey on Mechanistic Interpretability in AI

  2026cites this paper
• PyHealth 2.0: A Comprehensive Open-Source Toolkit for Accessible and Reproducible Clinical Deep Learning

  2026cites this paper
• Explainability Methods for Hardware Trojan Detection: A Systematic Comparison

  2026cites this paper
• Co-Designing Digital Humans for Online Learning: A Framework for Human-AI Pedagogical Integration

  2026cites this paper
• ProToken: Token-Level Attribution for Federated Large Language Models

  2026cites this paper
• A deep learning-based computational pipeline predicts developmental outcome in retinal organoids

  2026cites this paper
• A new era in identification of tick genera; artificial intelligence for precision and speed

  2026cites this paper
• The effect of whitening on explanation performance

  2026cites this paper