# VisualBackProp: efficient visualization of CNNs

arXiv: https://arxiv.org/abs/1611.05418 · HTML: https://mlsanity.com/paper/1611.05418/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Nov 2016.

## Built on by (most cited first, 10 of 10)

- [RES: A Robust Framework for Guiding Visual Explanation](https://mlsanity.com/paper/2206.13413/) (2022)
- [Adversarial Patch Attacks and Defences in Vision-Based Tasks: A Survey](https://mlsanity.com/paper/2206.08304/) (2022)
- [Towards Safe, Explainable, and Regulated Autonomous Driving](https://mlsanity.com/paper/2111.10518/) (2021)
- [Imitation Learning for End to End Vehicle Longitudinal Control with Forward Camera](https://mlsanity.com/paper/1812.05841/) (2018)
- [Learning with Chemical versus Electrical Synapses -- Does it Make a Difference?](https://mlsanity.com/paper/2401.08602/) (2023)
- [Imitation Learning via Simultaneous Optimization of Policies and Auxiliary Trajectories](https://mlsanity.com/paper/2105.03019/) (2021)
- [Multi-Abstractive Neural Controller: An Efficient Hierarchical Control Architecture for Interactive Driving](https://mlsanity.com/paper/2305.14797/) (2023)
- [Depth Matters: Multimodal RGB-D Perception for Robust Autonomous Agents](https://mlsanity.com/paper/2503.16711/) (2025)
- [HySafe-AI: Hybrid Safety Architectural Analysis Framework for AI Systems: A Case Study](https://mlsanity.com/paper/2507.17118/) (2025)
- [Synaptic Activation and Dual Liquid Dynamics for Interpretable Bio-Inspired Models](https://mlsanity.com/paper/2602.13017/) (2026)

## Abstract (arXiv)

This paper proposes a new method, that we call VisualBackProp, for visualizing which sets of pixels of the input image contribute most to the predictions made by the convolutional neural network (CNN). The method heavily hinges on exploring the intuition that the feature maps contain less and less irrelevant information to the prediction decision when moving deeper into the network. The technique we propose was developed as a debugging tool for CNN-based systems for steering self-driving cars and is therefore required to run in real-time, i.e. it was designed to require less computations than a forward propagation. This makes the presented visualization method a valuable debugging tool which can be easily used during both training and inference. We furthermore justify our approach with theoretical arguments and theoretically confirm that the proposed method identifies sets of input pixels, rather than individual pixels, that collaboratively contribute to the prediction. Our theoretical findings stand in agreement with the experimental results. The empirical evaluation shows the plausibility of the proposed approach on the road video data as well as in other applications and reveals that it compares favorably to the layer-wise relevance propagation approach, i.e. it obtains similar visualization results and simultaneously achieves order of magnitude speed-ups.