Activating Dance (AR)chives: Exploring Augmented Reality with the Repertoire of Martha Graham
Written by Hyunju Kim
The Challenge
Dance is a transient art form that encapsulates culture, tradition, and human expression. Therefore, preserving its essence—its emotional depth, cultural significance, and kinetic beauty—poses a significant challenge. Traditional archival methods, such as video recordings, often fall short of capturing the dynamic interactivity necessary to fully engage audiences, scholars, and dance students. These methods are particularly limited in capturing the fluidity and evolution of dance over time, resulting in gaps in how dance is preserved, studied, and experienced.
The choreographic works of Martha Graham, performed repeatedly over decades, have maintained their core elements as their emotional and expressive nuances evolve from generation to generation. This evolution presents a unique opportunity to address the challenges of dance preservation, and to explore innovative archival techniques that embrace the integrity of the art form while harnessing the affordances of new technologies.
The Discovery and Exploration Process
The Dance (AR)chive project emerged from a critical need to move beyond traditional methods of dance preservation. In particular, close collaboration with archivists, dancers, choreographers, and technologists proved essential for understanding the limitations of existing archival methods and identifying areas where augmented reality (AR) could bridge the gaps. Through experimentation with various technologies over the summer, we sought to determine effective methods for capturing, preserving, and activating dance in new ways. This iterative process led to an interrogation of digital archiving practices, with a particular focus on integrating existing archival data into a cohesive AR framework.
Generating 3D Dances from Existing Archival Videos
Recognizing that most archives are reliant upon video recordings – ‘flattened’ renderings of a live dance performance – we focused on utilizing standard camera footage to generate high-quality 3D models. These models were then transformed into virtual avatars, with animations replicating the original performances. Through interviews with dancers, we were encouraged to prioritize the accurate visualization of muscle and bone dynamics, essential for movement representation. To this end, simplified avatars (SMPL avatars) were employed, emphasizing the biomechanics of movement over customized or photorealistic representations that focus on appearance and identity. Moreover, the dancers responded positively to our process of taking archival footage and transmuting the performers into 3D representations, thereby inviting new forms of spatial dimensionality and interactivity. Not only could this approach enhance extant archival materials, it could have unique pedagogical implications for dancers wanting to learn Graham’s historic oeuvre.
Choreographic Search Engine
Observations of dance practices revealed that dancers often physically demonstrate movements when discussing specific choreographic elements. Additionally, many dance-related documents include images capturing dancers’ postures. In response, we developed a choreographic search engine, allowing users to search the dance archive using two forms of image input: (i) scanning an existing archival photograph or illustration, and (ii) scanning a live gesture or embodied pose. This unique tool facilitates more intuitive access to choreographic archives and is designed for seamless integration with existing archival data. Additionally, this resource allows teachers to easily identify where key elements of the Graham technique (e.g., the ‘release’ or ‘contraction’) appear across multiple works.
Performance Activations
The project further explored the potential of augmented reality, not only as a documentation tool but also as a medium for reactivating dance performances in new environments to enhance contextual, cultural, and historical dimensions. This exploration involved scanning specific physical spaces to create virtual replicas, and then integrating dance performances into different civic spaces seamlessly.
The Project
Building upon the previous discovery and exploration process, this project, entitled Dance (AR)chive, has been developed into three distinct applications, each tailored to a specific user experience:
XR Headset Application
This application allows users to explore the dance archive using XR headsets. Since Noguchi famously designed stage sets for Martha Graham’s productions, a 3D-scanned version of the museum in Queens invites users to be immersed within Noguchi’s sculptural world, and to have Graham’s work activated simultaneously. Users can also utilize real objects, such as dance photographs located in their personal spaces, to search for context-specific segments within the (AR)chive.
Trolley-Based Digital Mirror
Designed for dancers, this application features a movable digital mirror integrated within the (AR)chive. The trolley allows dancers to move freely and observe (AR)chived avatars alongside their own movements without the need for any wearable equipment. The mobility of the trolley enables dancers to change their perspective, offering multiple viewpoints for observing the avatars. Furthermore, dancers can physically demonstrate a specific movement and search to retrieve the corresponding choreographic moment from the archive.
Four Freedoms Park Exhibition
This application was specifically designed for use at Four Freedoms Park on Roosevelt Island. When visitors wear the XR headset, they can experience (AR)chive performances within the park itself. After reading the park’s declaration in stone to honor the ‘Freedom of Expression’, (AR)chived avatars appear and perform Graham’s“Ritual to the Sun” from Acts of Light (1981). In the original performance, dancers clad in shimmering suits symbolized the rising and setting of the sun. Drawing inspiration from this aesthetic framing, we activated the (AR)chive with avatars covered in reflective meshes, arranged in a group formation, similar to the original choreography. Activating Graham’s iconic work in different environments and contexts has the potential to foster new forms of historical connection and meaning.
Hyunju Kim
Ph.D. student, Information Science, Cornell University
Impact and Path Forward
The Dance (AR)chive project represents a significant advancement in the preservation and activation of dance through the innovative use of augmented reality. By addressing the limitations of traditional archival methods, this project has demonstrated how AR can capture the fluidity, spatiality, and evolving nature of dance. The project integrated existing archival data with immersive, interactive media to explore new ways for dance to be preserved, studied, and experienced. Looking ahead, there is potential to explore further how these AR-enhanced archives can be utilized by dancers and performers, particularly within the broader context of performance art. Particularly, we are interested in exploring how these archives can be connected to spaces and audiences in meaningful ways, enhancing the interactive experience for both performers and viewers. By continuing to develop and refine these techniques, we can ensure that the rich history and cultural significance of dance is not only preserved but also dynamically experienced in the present and future.