Explaining Motion Relevance for Activity Recognition in Video Deep Learning Models

Liam Hiley, Alun D. Preece, Yulia Hicks, Supriyo Chakraborty, Prudhvi Gurram, Richard TomsettarXiv 2020 · arXiv

Explaining Motion Relevance for Activity Recognition in Video Deep Learning Models

Introduction: The Challenge of Spatial Bias

3D Convolutional Neural Networks (CNNs) have become the standard for video-based activity recognition. However, these models often operate as "black boxes," making it difficult to discern whether they are identifying an action based on actual movement or merely relying on static scene context. For instance, a model might correctly classify a video as "tennis" simply by recognizing a tennis court, a phenomenon known as spatial bias.

While saliency maps—heatmaps indicating influential pixels—are commonly used for 2D image explanations, extending them to 3D video often conflates spatial and temporal information. This lack of granularity prevents researchers from determining if a model is truly learning the dynamics of an activity.

Methodology: Selective Relevance via Temporal Filtering

To address this, the authors propose selective relevance, a post-processing technique designed to isolate motion-specific information. This method acts as a temporal filter for existing explanation frameworks like Deep Taylor Decomposition (DTD).

The process follows three primary steps:

  1. Baseline Generation: A standard spatio-temporal explanation is generated for the 3D video input.
  2. Temporal Edge Detection: A Sobel operator is applied along the temporal ($t$) axis of the relevance map. This identifies regions where the model's attention shifts rapidly over time, serving as a proxy for motion.
  3. Masking: The resulting temporal derivatives are used to create a mask that filters out static background features, leaving only the "motion-relevant" components.
A step-by-step illustration of the selective relevance pipeline, showing how temporal derivatives isolate motion from static context.

Key Insights: Quantifying Contextual Dependency

A significant finding of this research is the extent of model bias toward static context. The authors quantified that for many baseline explanation methods, approximately 90% of the "relevant" information is actually static spatial context rather than motion. By stripping away this background "noise" (such as stadium seating or landscape features), the resulting explanations become significantly more interpretable for human users.

Furthermore, the study used optical flow as a ground truth for physical movement. The selective relevance method was found to significantly increase the alignment between model explanations and actual pixel movement, suggesting that the filtered maps provide a more honest representation of the model's "motion reasoning."

Results and Performance

The selective relevance approach demonstrates both quantitative and qualitative improvements:

  • Precision Gains: For the DTD method, "motion precision" (the overlap with optical flow) improved from 27.79% to 41.09%.
  • Efficiency: The method is highly efficient, adding only 0.5 to 1 millisecond of computational overhead, making it suitable for real-time model debugging.
  • Qualitative Clarity: Visual comparisons demonstrate that the method successfully isolates the primary actor. In activities like jet skiing or javelin throwing, irrelevant background elements like water spray or the track are removed, focusing the explanation on the athlete's movement.
Visual comparison demonstrating how selective relevance removes background noise across various activities, focusing on the moving subjects.

Limitations and Future Directions

Despite its success, the method faces challenges when paired with low-resolution explanation tools like GradCAM. Because GradCAM produces coarse heatmaps, the temporal changes are often too subtle for the Sobel filter to capture effectively. Additionally, the threshold for motion detection ($\sigma$) currently requires manual tuning per dataset.

Future work may involve applying this technique to modern architectures like D3D, which are specifically designed to emulate optical flow, to further bridge the gap between model explanations and physical reality.

Summary updated

Abstract

A small subset of explainability techniques developed initially for image recognition models has recently been applied for interpretability of 3D Convolutional Neural Network models in activity recognition tasks. Much like the models themselves, the techniques require little or no modification to be compatible with 3D inputs. However, these explanation techniques regard spatial and temporal information jointly. Therefore, using such explanation techniques, a user cannot explicitly distinguish the role of motion in a 3D model's decision. In fact, it has been shown that these models do not appropriately factor motion information into their decision. We propose a selective relevance method for adapting the 2D explanation techniques to provide motion-specific explanations, better aligning them with the human understanding of motion as conceptually separate from static spatial features. We demonstrate the utility of our method in conjunction with several widely-used 2D explanation methods, and show that it improves explanation selectivity for motion. Our results show that the selective relevance method can not only provide insight on the role played by motion in the model's decision -- in effect, revealing and quantifying the model's spatial bias -- but the method also simplifies the resulting explanations for human consumption.
BibTeX
@misc{hiley2020explaining,
  author = {Liam Hiley and Alun D. Preece and Yulia Hicks and Supriyo Chakraborty and Prudhvi Gurram and Richard Tomsett},
  title = {{Explaining Motion Relevance for Activity Recognition in Video Deep Learning Models}},
  year = {2020},
  eprint = {2003.14285},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2003.14285},
}
Figure