Floor sensors are cheap and easy to use! A Nihon Buyo Case Study

Miho Imai · 2025

Read Floor sensors are cheap and easy to use! A Nihon Buyo Case Study by Miho Imai (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-19261.txt.

Introduction

Pressure-sensing floors can track how dancers shift their weight, but the systems demand expert setup. Miho Imai tests Flexel, a modular, low-cost floor with fine pressure resolution, in Nihon Buyo, a traditional Japanese dance. A first-time, non-technical user installed and calibrated the floor, then ran it for nine weeks. The system recorded pressure data from a teacher and a learner, synced with video, while custom software processed the signals.

The expectation was that the learner's weight patterns would converge toward the teacher's. Instead, the numbers showed the learner settling into a consistent profile that stayed distinctly her own. Even inside a strict teaching tradition, an individual movement signature emerged. The finding concerns usability: the non-expert user deployed and operated the floor throughout. Imai argues that such cheap, easy sensing floors could spread into dance education, performance, and wider movement research.

Catalog record

CREATORMiho Imai
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE15 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-19261

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Floor sensors are cheap and easy to use! A Nihon Buyo Case Study Miho Imai Abstract . As floor-sensing technologies gain traction in movement research, questions remain about their usability and effectiveness for non-expert users. This study presents a case study evaluating Flexel, a modular, low-cost, high-resolution pressure-sensing floor interface, in the context of Nihon Buyo, a traditional Japanese dance. The system was installed, calibrated, and used by a first-time, non-technical user to track weight distribution patterns of a teacher and learner over nine weeks. Live pressure data was synchronized with video recordings, and custom software was developed to process and analyze the signal. Despite expectations that the learner’s weight distribution would converge toward the teacher’s over time, quantitative analyses revealed that the learner developed a consistent yet distinct movement profile. These findings suggest that even within rigid pedagogical structures, individual movement signatures can emerge. More importantly, the study demonstrates that Flexel can be deployed and operated effectively by non-expert users, highlighting its potential for broader adoption in education, performance, and embodied research. 1. Introduction As floor-sensing technologies gain traction in movement research, questions remain about their usability and effectiveness for non-expert users. We present a case study of high school students using Flexel ( yoshida2022flexel ) , a modular, low-cost, high-resolution pressure-sensing floor interface. The students used Flexel to understand the movement and progress of novice dancers in Nihon Buyo classes—a traditional Japanese dance. The students installed the system with minimal help, calibrated it, and used it as first-time non-technical users to track the weight distribution patterns of a teacher and a learner over a period of nine weeks. The students used Flexel to obtain pressure data in a simple CSV format, which they then synchronized to video recordings using external software. This allowed them to understand their results. Students were surprised: they had an initial hypothesis that a novice dancer’s weight distribution would, over time, align more closely to that of a professional instructor. But with the quantitative results that they obtained by using Flexel, they were able to discard this theory. Instead, they found that the novice dancer developed a consistent yet distinct movement profile. These findings suggest that even within rigid pedagogical structures, individual movement signatures can emerge. More importantly, this simple case study demonstrates that Flexel can be deployed and operated effectively by non-expert users, highlighting its potential for broader adoption in education, performance, and embodied research. 2. Background Below we provide some background on Flexel and Nihon Buyo. 2.1. Flexel Figure 1 . Nihon Buyo professional dancer standing on a deployment of Flexel ( yoshida2022flexel ) . Flexel ( yoshida2022flexel ) is a floor system that can track where people are standing, walking, or moving. Instead of using cameras or wearables, Flexel uses special tiles with built-in pressure sensors that feel how much weight is on them and where it is coming from. Each tile, covering 0.25m^{2} , is made up of 9 sensing units and can be easily placed next to other tiles to cover any size room. Flexel can maintain high sensing accuracy while reducing sensor density, achieving comparable performance with only 1/81 the number of the density of sensors typically required by commercial pressure mats. This makes it cheaper and easier to set up. 2.2. Nihon Buyo Nihon Buyo ( malm1977legacy ) is a classical Japanese dance form that is structured and requires precise, controlled movements. Dancers are traditionally trained to follow the steps of a master, spending years as apprentices to learn precise moves. Because of the strict nature of the dance, one might wonder whether dancers end up moving the same way, or if they eventually develop their own unique style over time. Figure 2 . High level overview of the experiment that we performed. It includes asking an instructor and a learner to independently practice a dance and then perform it on top of a Flexel deployment multiple times throughout 9 weeks. The data is recorded an analyzed over theis period of time. 3. Objective In this project we report on the experience of high school students who used Flexel to explore a burning question they had: in Nihon Buyo, does the weight distribution of novice dancers start to replicate the weight distribution of an expert instructor over time? By collecting and analyzing data using Flexel, the students were able to collect evidence for and against this hypothesis. 4. Methodology Figure 3 . Screenshot of the software that analyzes the data collected by Flexel. The software displays the pressure and movement in a variety of visual formats that are easy for users to see and understand. Measurement process. The students measured the weight pressure distribution of dancers with Flexel. Flexel can sample the pressure distribution at 80Hz. In every square meter of floorboard, there are 144 load cells installed. The students used 4 of these floorboards to capture the weight pressure distributions of the dancers, creating an area of four square meters for each dancer. The students then connected Flexel onto a PC and opened up its collection software (“Unisession for Floor”). At this point, Flexel would stream measurement data and the software would report the pressure distribution and display a pressure image for the students to see and analyze. During this process, Flexel automatically applied a calibration algorithm to ensure accurate and reliable measurements. At the end of the measurement, the software produced a CSV file with the data. Collecting dance data. To create the teacher’s dance dataset, the students recorded a professional dancer performing a dance called “Seven Lucky Gods” 3 different times. They then asked a novice dancer to practice the same dance for a period of 9 weeks. The novice dancer had access to a video recording of the teacher’s dance during this timeframe. Each week, the students measured the novice dancer’s movements using Flexel 3 times. The students stopped collection after 9 weeks, because the instructor deemed that the novice dancer had fully learned the dance. Processing the data The students processed the data using a Jupyter Notebook. They first determined the x and y center of mass of the dancer by finding the average of the x-axis and y-axis and locating a point with those x and y values recorded on the floorboards. Since the students did not have background in statistics or data science, an expert provided them with an equation to calculate the center of pressure (COP) in both the x and y directions, which they did on the Jupyter Notebook. The equations were as follows: (1) x_{COP}=\frac{\sum_{i}x_{i}F_{i}}{\sum_{i}F_{i}} (2) y_{COP}=\frac{\sum_{i}y_{i}F_{i}}{\sum_{i}F_{i}} The COP represents the average location of all forces acting on the surface, essentially indicating where the body’s weight is concentrated. In the equations, x_{i} and y_{i} are the positions where force is applied, while F_{i} represents the amount of force at each point. The total force, F_{total} , is the sum of all these individual forces. By taking the weighted average of these positions, the COP helps analyze balance and weight distribution, which is crucial in studying movement control. In the figures, the COP sensor data is represented in pixel units of floor sensor. To convert the pixel values into centimeters, they should be multiplied by 8.3, as there

Read the complete text