Exploring the Creative Possibilities of Infinite Photogrammetry through Spatial Computing and Extended Reality with Wave Function Collapse Aviv Elora , Samantha Condea a University of California, Santa Cruz, Department of Computational Media, Jack Baskin School of Engineering, 1156 High St, Santa Cruz, California, 95064, United States Abstract Modern extended reality systems that merge virtual and augmented reality provide a unique design space for creative applications. These devices have begun to incorporate spatial computing, or methods of runtime digital photogrammetry which translate the physical world into the virtual. In this study, we examine the use of extended reality for “infinite photogrammetry,” a system of mapping the physical world into a virtual experience and procedurally generating an infinite version of the scanned architecture. We explore our system through a use case of mapping a residential home for infinite photogrammetry with the Magic Leap Spatial Computing Headset, Wave Function Collapse Algorithm, and Unity Game Engine. We conclude with a discussion on the creative applications of infinite photogrammetry and considerations for future research. Keywords Infinite Photogrammetry, Photogrammetry, Spatial Computing, Extended Reality, Virtual Reality, Aug- mented Reality, Wave Function Collapse, Procedural Content Generation, Applied Generative Algorithms 1. Introduction Modern extended reality (XR) systems have gone a long way technologically in enhancing user immersion through widening the field of view, increasing frame-rate, leveraging low latency motion capture, and providing realistic surround sound [1]. As a result, we see a new wave of mass adoption of commercial XR Head-Mounted Displays (HMDs) such as the Magic Leap One, Microsoft Hololens, HTC Vive, Oculus Quest, PlayStation Morpheus, and more that have entered the market with nearly over 200 million projected systems sold since 2016 [1]. These systems are becoming ever more mobile and intrinsic to the average consumer’s entertainment experience, enabling a mode of full-body engagement combining the physical and virtual world [2, 3]. More recently, these devices have begun incorporating simultaneous localization and mapping to transfer the physical world’s architecture into the digital environment, as seen with the photogrammetry like spatial computing and meshing capabilities of the Magic Leap One [4]. These mediums provide new opportunities to explore tools for casual creation and generative Joint Proceedings of the ICCC 2020 Workshops (ICCC-WS 2020), September 7-11 2020, Coimbra (PT) / Online Envelope-Open aelor@ucsc.edu (A. Elor); sconde@ucsc.edu (S. Conde) GLOBE https://www.avivelor.com/ (A. Elor); https://samanthaconde.cargo.site (S. Conde) Orcid 0000-0001-5356-3948 (A. Elor) © 2020 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR Workshop Proceedings http://ceur-ws.org ISSN 1613-0073 CEUR Workshop Proceedings (CEUR-WS.org) computing. The use of XR and photogrammetry have been increasing due to the benefits that result from this combination. The main usage of this combination has been to reconstruct objects or locations from the real world to a mixed reality environment. Virtual reality (VR) has been getting most of the attention for recreating real-life objects and locations, but what is not mentioned is the time it takes to develop these things. VR usually takes a lot more time, precision, and accuracy to develop an object as compared to AR. Portalés et al. have found that utilizing Augmented Reality (AR) with photogrammetry has been more cost-efficient to create such objects [5]. Not only that, but the time saved using AR with photogrammetry is almost more than 50% [5]. AR and photogrammetry have their advantages not only for being more time and cost-efficient but also for providing accessibility to people. Two examples where AR and photogrammetry were utilized to create more accessible environments are Drap et al.’s VENUS project and Pietroszek’s mixed reality exhibition. Drap et al. used photogrammetry to survey marine areas of the Pianosoa island. It is a step forward to having archaeologists investigate untouched and unreachable areas of the deep ocean [6]. This is a great way to digitally archive and preserve underwater findings without compromising them. In a related application, Pietroszek created a mixed reality exhibition to make it more accessible for people who cannot visit a normal exhibition due to location, disability, or socioeconomic status [7]. From these works, we argue that the incorporation of extended reality devices may provide unique opportunities for casual recreation. In 2015, Compton & Mateas defined an alternative design space for system creation: “A Casual Creator is an interactive system that encourages the fast, confident, and pleasurable exploration of a possibility space, resulting in the creation or discovery of surprising new artifacts that bring feelings of pride, ownership, and creativity to the users that make them” [8]. These tools emphasize creativity and design support by enabling a flow of choice and rapid iteration while providing both passive and active automation [9]. Moreover, the curious users of casual creators have been hypothesized to be driven primarily by the curiosity and capability of a system’s design space [10]. In this study, we explore the usage of an XR headset to understand the potential of these devices for the use of photogrammetry, converting the physical world into the virtual. We also examine autonomy for XR enabled photogrammetry to explore how it can be extended for generative experiences through examining procedural content generation (PCG). PCG algorithms applied to photogrammetry may produce some interesting design artifacts for game and experience design. As games have been evolving rapidly, so has the use of PCG [11]. Designers use PCG to implement content that has been automatically generated from assets at random [12]. In this definition, content is a broad term for what researchers, game designers, and academics would want to generate. Applying PCG to photogrammetry helps game designers create infinite possibilities for levels, non-playable characters, and many more objects in a digital game. This combination allows for more opportunities to surprise users and even the designers themselves. An algorithm for PCG that has been gaining traction in the creative design world is Gumin’s WaveFunctionCollapse (WFC) [13]. WFC is a non-backtracking, greedy search algorithm that enables large output generated from a small number of constraints determined by a window of input media. The algorithm has attracted the attention of game creators, PCG researchers, and level designers over the past years [14, 15]. It enables designers to speed up the time and production costs of asset creation while providing them with constraints to manipulate pattern generation. For our study, we are interested in extending this algorithm to 3D world generation by utilizing photogrammetry with an extended reality headset. To the best of our knowledge, this study is one of the first to bridge spatial computing with WFC for Infinite Photogrammetry. We hope to examine the combination of these technologies to demonstrate a proof of concept and consider its creative applications for future research. (a) Scanning the Physical World (b) Infinite Photogrammetry Pipeline Figure 1: On the left we see a user walking around their home with a Magic Leap One. House geometry is procedurally meshed in the unity game engine and stored as mesh for manipulation with Wave Function Collapse. On the right we see the system pipeline for creating infinite photogrammetry. 2. System Design This project leverages the capabilities of the Magic Leap Spatial Computing Headset when combined with the Wave Function Collapse Algorithm and the Unity Game Engine. The goal was to create a playable experience that generates infinite photogrammetry of a scanned environment. To this end, we designed our system to (1) create a methodology of translating physical world geometry into virtual 3D environments with Magic Leap, (2) adapt a prior Wave Function Collapse algorithm to generate new architecture via Unity3D, and (3) to explore the application of infinite photogrammetry. This process can be described in four stages: capturing, meshing, formatting, and building, as shown in Figure 1b. In this section, we discuss the tools we used to enable this system’s design. The Magic Leap One (MLO) headset, an extended reality interaction system, is a “spatial computing” headset that overlays augmented reality while performing simultaneous localization and mapping on the physical world [4]. MLO was examined as a development platform because, at the time of this study, little to no evaluations were found in academia for development testing of our proposed application. Seeing the physical world around the user is critical for safety when mapping environments. The untethered headset differs from other commercially available XR HMDs by projecting light directly into the user’s eyes while also enabling higher input modalities through hand tracking, eye tracking, dynamic sound fields, and 6-Degree of Freedom (DoF) controllers with haptic feedback [4]. To enable the visualization and interaction with the virtual world, the Unity Game Engine was chosen as the primary driver of our experience. Unity is a flexible real-time 3D development platform that enables the creation, operation, and rapid prototyping of interactive virtual content [16]. Unity was chosen due to its flexible capabilities, which allows it to build the same experience between multiple operating systems such as WebGL, Magic Leap, HTC Vive, Windows, Mac, and more [16]. Thus, we developed our experience in Unity 2019.1.5f1 through two separate build instances: Lumin (MLO SDK 0.21) and WebGL (OpenGL 4.5). To obtain a mesh of the physical world, we utilized the MLO World Reconstruction Spatial Mapper, an algorithm to detect real-world surfaces and construct a runtime virtual mesh to represent the real world’s collisions for game engine [17, 18]. We converted the world reconstruction mapper into a serialized mesh during runtime, which is then stored as an asset for later manipulation. This process allows us to capture the rough geometry of the user’s surroundings as they walk through and map their desired game architecture, as shown in Figure 1a. From there, we translate the asset into a playable scene to allow the user to walk through and navigate their scans virtually. We examined this process during a ten-minute session as a user walked through their home. This consisted of rapidly scanning a 1459 square foot residential home with two bedrooms, two bathrooms, one office, a living room, and a kitchen. The results of this process can be seen in Figure 1a and 2a, where some of the rooms are reconstructed for the user to virtually walk around their scans in the unity game engine. After the scanned geometry is captured and serialized to independent mesh assets, we then proceed to format the assets for WFC. To enable PCG, we modified Kleineberg’s Infinite City adaption of WFC [19]. Using the serialized meshes of the geometry scanned by the MLO spatial mapper, we divide the rooms into one-meter voxels and define WFC constraints through mapping the six sides of the room with numbered adjacency keys as shown in Figure 2b. The user is then able to define the WFC generative adjacency of rooms through one-meter voxels chunks. Such rooms become generated through chunks in relation to the user’s world position in the unity engine. As a result of this process, we end with a unity experience that can generate an infinite form of photogrammetry produced from the MLO Mixed Reality headset. The infinite house produced from this process can be seen in Figure 2b. A demo of the experience can be found at https: // github.com/ avivelor/ InfinitePhotogrammetry. 3. Results and Discussion We were able to successfully test our system in a residential home and generate an infinite version of the house from a ten-minute scanning session. This produced a virtual experience in which the user was able to re-visit the scanned geometry and walk through both a static and an infinite WFC generated version of the home. Subsequently, our exploratory system may suggest that utilizing the spatial computing capabilities of modern XR devices can produce interesting virtual artifacts from both a static and generative perspective. In this section, we reflect on our system design for creative use and consider future research areas to understand how Infinite Photogrammetry could be better tailored as a creative tool. Spatial computing systems are becoming ever more mainstream with consumer applications such as Snapchat, Instagram, and Facebook, who leverage augmented reality filters for social communication in videos and photos [20]. As XR devices become more affordable, we may see a similar trend in this adoption and should consider the creative possibilities of XR’s enhanced input modalities and full-body interaction. More creative applications, such as Minecraft Earth, are beginning to utilize AR for users to build block-based game worlds within their own homes [21]. Other researchers are exploring extended reality for creative tools within architecture, art, design, games, media, and e-publishing [22]. This includes extended reality creator tools such as collaboration and education [23, 24]. Such tools and environments have been shown to positively impact mental health [25], learning [26], and physical exercise [27, 28]. (a) Static House (b) Infinite House Figure 2: On the left we see examples of stored mesh geometry from a user during a ten-minute scanning session. A custom depth shader is applied and overlaid over the physical world to represent scanned geometry for the user with distance correlating to color. The user can review the scans by walking through the house in the unity engine through an MLO headset, a standalone build, or a WebGL build. On the right we see An infinite house generated by the users scanned geometry and wave function collapse. The user is able to define the WFC generative adjacency of rooms through chunks of the scanned geometry represented by one meter voxels. The rooms are then generated through chunks in relation to the users world position in the unity engine. For our proposed Infinite Photogrammetry application, more work must be done to determine its creative possibilities and refine its uses toward a casual creator. More evaluation must be done with Infinite Photogrammetry on more architectures such as museums, outdoor parks, and historical sites. In addition, efforts must be made to increase understanding of user perception and creativity within the tool. From this end, we believe that Infinite Photogrammetry may be of interest to be explored within the following fields: • Video Game Designers interested in mapping real-world architecture for generative or static game levels; • Artists of virtual environments interested in emergent design patterns from real-world terrain; • Film producers scouting physical locations for filmmaking and or capturing virtual assets for special effects; • and curious creators interested in exploring the design space of infinite photogrammetry for world-building and manipulation. To this end, Infinite Photogrammetry may enable a system of creators to capture real-world environments with ease and creatively manipulate them from both static and PCG perspectives. We hope to refine this system for multiple extended reality devices such as mobile augmented reality with ARKit, ARCore, and WebXR [29, 30]. Additionally, it may be interesting to influence infinite photogrammetry with emotion personalization, which can be tuned from an immersive virtual environment [31]. More significant input systems should be crafted and explored to enabled runtime creator tools such as manipulating WFC adjacency, smoothing scanned world geometry, and translating base color texture from world reconstruction. 4. Conclusion In this paper, we presented the creative application of Infinite Photogrammetry. We discuss how modern extended reality headsets can be utilized for Infinite Photogrammetry to translate the physical world into a virtual environment. We piloted our system through scanning a residential home to transfer a user’s surrounding into a playable experience that can be infinitely generated with the Wave Function Collapse algorithm. Lastly, we considered the creative possibilities of this application as well as areas for future research. Although more work is to be done, a step towards Infinite Photogrammetry may enable a deeper dive into the creative manipulation of the physical world through the virtual. Acknowledgments The authors would like to thank and acknowledge Professor Angus Forbes for his advice and expert opinion during the exploration of this project. References [1] M. Beccue, C. Wheelock, Research Report: Virtual Reality for Consumer Markets, Technical Report, Tractica Research, 2016. URL: https://www.tractica.com/research/ virtual-reality-for-consumer-markets/. [2] S.-N. Chang, W.-L. Chen, Does visualize industries matter? a technology foresight of global virtual reality and augmented reality industry, in: 2017 International Conference on Applied System Innovation (ICASI), IEEE, 2017, pp. 382–385. [3] S. Liu, Forecast Augmented (AR) and Virtual Reality (VR) Market Size Worldwide From 2016 to 2023 (in Billion US Dollars), Statista, 2019. [4] M. Leap, Magic leap one–creator edition, Internet: https://www. magicleap. com/magic- leap-one [Jan. 19, 2019] (2019). [5] C. Portalés, J. L. Lerma, S. Navarro, Augmented reality and photogrammetry: A synergy to visualize physical and virtual city environments, ISPRS Journal of Photogrammetry and Remote Sensing 65 (2010) 134–142. [6] P. Drap, J. Seinturier, D. Scaradozzi, P. Gambogi, L. Long, F. Gauch, Photogrammetry for virtual exploration of underwater archeological sites, in: Proceedings of the 21st international symposium, CIPA, 2007, p. 1e6. [7] K. Pietroszek, Mixed-reality exhibition for museum of peace corps experiences using ahmed toolset, in: Symposium on Spatial User Interaction, 2019, pp. 1–2. [8] K. Compton, M. Mateas, Casual creators., in: ICCC, 2015, pp. 228–235. [9] K. Compton, Casual creators: Defining a genre of autotelic creativity support systems, University of California, Santa Cruz, 2019. [10] M. J. Nelson, S. E. Gaudl, S. Colton, S. Deterding, Curious users of casual creators, in: Proceedings of the 13th International Conference on the Foundations of Digital Games, 2018, pp. 1–6. [11] M. Hendrikx, S. Meijer, J. Van Der Velden, A. Iosup, Procedural content generation for games: A survey, ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) 9 (2013) 1–22. [12] J. Togelius, E. Kastbjerg, D. Schedl, G. N. Yannakakis, What is procedural content gener- ation? mario on the borderline, in: Proceedings of the 2nd international workshop on procedural content generation in games, 2011, pp. 1–6. [13] M. Gumin, Wavefunctioncollapse, GitHub repository (2016). [14] I. Karth, A. M. Smith, Wavefunctioncollapse is constraint solving in the wild, in: Proceed- ings of the 12th International Conference on the Foundations of Digital Games, 2017, pp. 1–10. [15] A. Sandhu, Z. Chen, J. McCoy, Enhancing wave function collapse with design-level constraints, in: Proceedings of the 14th International Conference on the Foundations of Digital Games, 2019, pp. 1–9. [16] Unity Technologies, Unity real-time development platform | 3d, 2d vr & ar, Internet: https://unity.com/ [Jun. 06, 2019] (2019). [17] M. Leap, Magic leap developer - spatial meshing, Internet: https://developer.magi- cleap.com/en-us/learn/guides/meshing-in-unity [May. 29, 2020] (2019). [18] D. DeTone, T. Malisiewicz, A. Rabinovich, Toward geometric deep slam, arXiv preprint arXiv:1707.07410 (2017). [19] M. Kleineberg, Infinite procedurally generated city with the wave function collapse algorithm, Internet: https://marian42.de/article/wfc/ [May. 29, 2020] (2019). [20] D. Harborth, Augmented reality in information systems research: a systematic literature review, in: Twenty-third Americas Conference on Information Systems, Boston, 2017. [21] S. Khanna, Augmented reality: The present and the future, CYBERNOMICS 1 (2019) 15–18. [22] M. Abbasi, P. Vassilopoulou, L. Stergioulas, Technology roadmap for the creative industries, Creative Industries Journal 10 (2017) 40–58. [23] D. Andone, M. Frydenberg, Experiences in online collaborative learning with augmented reality., eLearning & Software for Education 2 (2017). [24] S. Serafin, A. Adjorlu, N. Nilsson, L. Thomsen, R. Nordahl, Considerations on the use of virtual and augmented reality technologies in music education, in: 2017 IEEE Virtual Reality Workshop on K-12 Embodied Learning through Virtual & Augmented Reality (KELVAR), IEEE, 2017, pp. 1–4. [25] D. Potts, K. Loveys, H. Ha, S. Huang, M. Billinghurst, E. Broadbent, Zeng: Ar neurofeedback for meditative mixed reality, in: Proceedings of the 2019 on Creativity and Cognition, 2019, pp. 583–590. [26] G. Papanastasiou, A. Drigas, C. Skianis, M. Lytras, E. Papanastasiou, Virtual and augmented reality effects on k-12, higher and tertiary education students’ twenty-first century skills, Virtual Reality 23 (2019) 425–436. [27] K. Kunze, K. Minamizawa, S. Lukosch, M. Inami, J. Rekimoto, Superhuman sports: Applying human augmentation to physical exercise, IEEE Pervasive Computing 16 (2017) 14–17. [28] A. Elor, M. Teodorescu, S. Kurniawan, Project star catcher: A novel immersive virtual reality experience for upper limb rehabilitation, ACM Transactions on Accessible Computing (TACCESS) 11 (2018) 1–25. [29] J. Linowes, K. Babilinski, Augmented Reality for Developers: Build practical augmented reality applications with Unity, ARCore, ARKit, and Vuforia, Packt Publishing Ltd, 2017. [30] B. Maclntyre, T. F. Smith, Thoughts on the future of webxr and the immersive web, in: 2018 IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR- Adjunct), IEEE, 2018, pp. 338–342. [31] A. Elor, A. Song, isam: Personalizing an artificial intelligence model for emotion with pleasure-arousal-dominance in immersive virtual reality, in: 2020 15th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2020)(FG), IEEE, 2020, pp. 583–587.