# Feature Visualization in 3D Convolutional Neural Networks

arXiv: https://arxiv.org/abs/2505.07387 · HTML: https://mlsanity.com/paper/2505.07387/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 22 references that are in the map.
- First posted on arXiv 12 May 2025.

## Builds on (most cited first, 11 of 11)

- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [Learning Deep Features for Discriminative Localization](https://mlsanity.com/paper/1512.04150/) (2015)
- [RAFT: Recurrent All-Pairs Field Transforms for Optical Flow](https://mlsanity.com/paper/2003.12039/) (2020)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [Deformable Convolutional Networks](https://mlsanity.com/paper/1703.06211/) (2017)
- [Striving for Simplicity: The All Convolutional Net](https://mlsanity.com/paper/1412.6806/) (2014)
- [Open-Sora: Democratizing Efficient Video Production for All](https://mlsanity.com/paper/2412.20404/) (2024)
- [Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks](https://mlsanity.com/paper/1910.01279/) (2019)
- [Understanding Neural Networks Through Deep Visualization](https://mlsanity.com/paper/1506.06579/) (2015)
- [Explainable 3D Convolutional Neural Networks by Learning Temporal Transformations](https://mlsanity.com/paper/2006.15983/) (2020)

## Abstract (arXiv)

Understanding the computations of convolutional neural networks requires effective visualization of their kernels. While maximal activation methods have proven successful in highlighting the preferred features of 2D convolutional kernels, directly applying these techniques to 3D convolutions often leads to uninterpretable results due to the higher dimensionality and complexity of 3D features. To address this challenge, we propose a novel visualization approach for 3D convolutional kernels that disentangles their texture and motion preferences. Our method begins with a data-driven decomposition of the optimal input that maximally activates a given kernel. We then introduce a two-stage optimization strategy to extract distinct texture and motion components from this input. Applying our approach to visualize kernels at various depths of several pre-trained models, we find that the resulting visualizations--particularly those capturing motion--clearly reveal the preferred dynamic patterns encoded by 3D kernels. These results demonstrate the effectiveness of our method in providing interpretable insights into 3D convolutional operations. Code is available at https://github.com/YatangLiLab/3DKernelVisualizer.