Deep within the tropical forests of central Java, Indonesia, surrounded by volcanoes near the equator, stands the monumental stone temple of Borobudur (Fig. 1). When we first encountered this structure, we were overwhelmed by its massive scale and the stark appearance of its dark volcanic stone—so different from traditional Japanese Buddhist architecture. At the same time, however, we felt that there might be a common thread between this temple and structures such as the Great Buddha of Nara, built in Japan in the 8th century: both embody humanity’s desire to seek spiritual support and salvation in something vast and transcendent.

Figure 1. Overview of Borobudur Temple

Constructed around the 9th century, Borobudur is one of the largest Buddhist monuments in the world and is now designated as a UNESCO World Heritage Site. The base forms a square approximately 120 meters on each side, and the structure rises to a height of about 35 meters. The monument is essentially a massive stupa without an internal chamber, composed of a nine-tiered stepped structure. Each level contains corridors lined with carved wall reliefs (Fig. 2). These layered stone corridors hold approximately 2,600 relief panels, through which visitors historically walked in a processional path, learning Buddhist teachings. The narratives depicted on the reliefs are arranged so that as one ascends the structure, one symbolically progresses toward enlightenment.

Figure 2. Stupas on the uppermost terrace (left) and examples of wall reliefs (center and right)

As computer science researchers, we have been working on recording Borobudur using 3D measurement techniques and reconstructing its structure in digital space. By analyzing large-scale point cloud data obtained through laser scanning and photogrammetry, we aim to support the preservation and understanding of cultural heritage (an example of such analysis is shown in Fig. 3). Our research seeks to construct a digital twin of the temple that includes not only its exterior but also its internal and subsurface structures (Fig. 4). While the external structure is precisely captured using 3D measurement techniques, the underground structure is reconstructed based on data from UNESCO drilling surveys. Furthermore, for the so-called “hidden foot” relief panels—currently concealed beneath later stone coverings—we are developing methods to reconstruct their 3D geometry using AI techniques applied to historical photographs.

Figure 3. 3D shape analysis of the Borobudur Ship relief

Figure 4. Transparent visualization of Borobudur Temple (including the hidden foot and subsurface foundation structure)

Using this digital twin, it is possible to visualize internal structures through transparent rendering and to experience the entire monument within a virtual reality (VR) environment. This represents an attempt to reinterpret a thousand-year-old stone monument through the lens of cutting-edge information technology.

3D measurement is not merely a tool for cultural preservation; it is also a means of reinterpreting information encoded in stone in a new digital form. When the entire structure and its details are examined computationally, features that are difficult to perceive on site may become visible. For example, a hypothesis proposed by Indonesian researchers suggests that the slight asymmetry of a Buddha statue’s ears may have been intentionally designed to interact with moonlight. Such insights are made possible through high-precision digital data. In this sense, digital twins provide a new methodological framework for understanding cultural heritage.

However, this research did not proceed smoothly. Fieldwork in Indonesia presented numerous challenges. The tropical climate posed a major obstacle: under intense sunlight, equipment temperatures rise rapidly, and camera failures are not uncommon. During the dry season, heavy dust requires extreme care in handling sensitive instruments. In the rainy season, sudden downpours frequently interrupt outdoor measurements. In fact, repeated equipment failures due to heat sometimes forced us to revise our plans significantly.

Collaboration with local staff was also essential, but differences in culture and daily practices sometimes had unexpected effects. For example, during the Islamic fasting month of Ramadan, many collaborators were unable to participate in daytime work. Research schedules had to be adjusted flexibly in response to such cultural and religious factors.

Perhaps the greatest challenge was obtaining permission to conduct measurements at a World Heritage site. In Indonesia, administrative procedures can be both complex and influenced by personal relationships. For instance, when the director of the heritage conservation office overseeing Borobudur changed, previously ongoing approval processes were effectively reset. As a result, it took several years to obtain official permission. These conditions placed considerable stress on younger researchers participating in the project. Despite traveling to Indonesia after long preparation, there were times when measurements could not proceed at all due to administrative issues. At one point, a junior researcher even expressed frustration openly. Managing such situations and maintaining good relationships with local stakeholders became an essential part of our role as project leaders. This experience reinforced our understanding that international collaborative research involves not only technology but also the careful management of human relationships.

The 3D data acquired under such challenging conditions now serve as invaluable resources for understanding the structure of Borobudur. Using the digital twin, we can observe the entire structure from a global perspective, examine internal structures through transparent visualization, and even explore the monument from viewpoints that are impossible in reality within a VR environment. A stone monument built over a thousand years ago is thus reinterpreted through modern information technology.

This research is conducted as an international collaborative project centered at Ritsumeikan University, in cooperation with Indonesia’s National Research and Innovation Agency (BRIN) and the Borobudur Museum. It also benefits from technical support from the Nara National Research Institute for Cultural Properties, which has extensive experience in archaeological measurement and preservation. Although Borobudur has been the subject of many studies, this project represents the first attempt by an international research collaboration to construct a comprehensive digital archive of the entire monument.

The corridors of Borobudur contain numerous relief panels depicting Buddhist narratives, as well as scenes of everyday life in ancient Indonesia. Among these are detailed carvings of large sailing vessels that once traveled between the Indonesian islands, now known as the “Borobudur ships” (Figs. 2 and 3). These depictions provide valuable insights into historical maritime trade and navigation technologies. The reliefs also portray clothing, dwellings, and social practices, offering a vivid picture of 9th-century Indonesian society.

Digitally preserving these reliefs through 3D measurement is not merely about recording their shapes; it is about preserving the historical and cultural information embedded within them for future generations. It is our hope that this research, conducted in a distant region of Java, will contribute in some small way to a deeper global understanding of cultural heritage.

作者简介

Satoshi Tanaka（田中觉教授）

Vice President of the Asia Simulation Federation, College of Information Science and Engineering, Ritsumeikan University（亚洲仿真联盟副理事长，立命馆大学信息科学与工程学院）

Liang Li（李亮教授）

College of Information Science and Engineering, Ritsumeikan University（立命馆大学信息科学与工程学院）