Abstract
Compact visual summaries of spatio-temporal movement data often strive to express accurate positions of movers. We present SpatialRugs, a technique to enhance the spatial awareness of movements in dense pixel visualizations. SpatialRugs apply 2D colormaps to visualize location mapped to a juxtaposed display. We explore the effect of various colormaps discussing perceptual limitations and introduce a custom color-smoothing method to mitigate distorted patterns of collective movement behavior.
Introduction
Analyzing the movement of large groups—such as schools of fish, pedestrian crowds, or vehicle fleets—presents a significant scalability challenge. Traditional trajectory plots often result in a "spaghetti" mess of overlapping lines, while animations place a heavy cognitive load on the viewer to track long-term trends.
A previous technique, MotionRugs, addressed density by representing each mover as a single pixel in a 1D vertical stack over time. However, this approach relies on spatial linearization, which strips away the physical context of where entities are located. SpatialRugs bridges this gap by providing spatial awareness within the space-efficient "rug" format.
Spatial Mapping via 2D Colormaps
SpatialRugs maps the (x, y) coordinates of each mover to a 2D color space. By anchoring specific colors to spatial regions (e.g., top-right as green, bottom-left as red), a collective color shift in the visualization immediately reveals the group's trajectory. For instance, using the Ziegler map allows users to intuitively associate primary colors with cardinal directions.
Time-Aware Color Smoothing (TACS)
Mapping space to color often creates "visual outliers"—single pixels that jump to a different color because of minor movements across color-space boundaries. Standard Gaussian blur makes the whole visualization fuzzy and unreadable. TACS addresses this by using a spatio-temporal "pooling matrix" that looks at a pixel’s neighbors and its future positions. By calculating the color median, TACS removes noise while keeping the overall movement patterns crisp and structurally sound.
Results and Evaluation
The technique was evaluated using a dataset of 151 fish moving in a tank for approximately 90 seconds. Qualitative results showed that complex maneuvers, such as a school turning from the upper right to the lower left, were clearly visible as distinct color transitions in the rug.
Quantitatively, TACS was compared against standard Gaussian blur using metrics like Root Mean Squared Error (RMSE) and Structural Similarity Index (SSIM). TACS consistently outperformed Gaussian blur in preserving data structure. For example, a TACS [15,15] configuration achieved an SSIM of 0.524, compared to 0.505 for a comparable Gaussian blur, demonstrating its superior ability to reduce noise without losing critical visual information.
Conclusion
SpatialRugs provides an intuitive overview of collective movement, though it is intended for qualitative trend analysis rather than high-precision coordinate tracking. Future work includes developing adaptive colormaps that respond to mover density and interactive anchors to link the visualization directly to physical maps.
Summary updated