Since its inception in the early 1980s, FPP has been fundamentally built upon geometric triangulation. By projecting structured fringe patterns onto an object and analyzing the resulting phase information, the system reconstructs the three-dimensional shape of the target. This framework has enabled remarkable advances in industrial metrology, intelligent manufacturing, and scientific research.
However, conventional FPP relies on the assumption that the captured signal mainly originates from direct surface reflections. In real-world scenarios involving highly reflective materials, translucent media, biological tissues, and complex environments, light propagation becomes significantly more complicated due to multiple reflections, subsurface scattering, and other global illumination effects. Under such conditions, the recorded measurements contain not only geometric information but also rich information about the underlying light transport process.
The authors argue that future 3D vision systems must answer a fundamentally different question. Rather than asking simply “Where is the object?”, the next challenge is understanding “How does light propagate through the scene?”. This shift extends the focus of 3D vision from geometric reconstruction to light transport analysis, marking the transition from measuring 3D to understanding 3D.
According to the review, the evolution of FPP can be divided into three stages: the Foundation Phase (1983–2006), the Booming Phase (2007–2018), and the Transformative Phase (2019–present) as shown in Fig. 2. The foundation phase established the theoretical framework of FPP, while the booming phase focused on improving measurement accuracy, speed, and hardware performance, driving the technology from laboratory research to widespread industrial deployment.
Today, the field is entering a transformative stage. As 3D sensing applications expand toward extreme scales, dynamic scenes, diverse materials, and challenging environments, many longstanding limitations can no longer be overcome simply through incremental performance improvements. Framework innovation is becoming the new driving force.
In this context, artificial intelligence (AI) and computational imaging (CI) are opening unprecedented opportunities. AI offers powerful capabilities for solving complex inverse problems, while CI introduces higher-dimensional physical models to describe light–matter interactions. Their convergence is driving FPP beyond traditional geometric triangulation toward a new computational 3D imaging framework.
Despite these advances, several fundamental challenges remain as shown in Fig. 3. The first is extending measurement capabilities from ideal diffuse surfaces to real-world scenes containing specular reflections, inter-reflections, subsurface scattering, and transparent materials. The second is moving beyond geometric reconstruction toward light transport analysis in environments involving strong ambient illumination, scattering media, underwater imaging, and multipath propagation. The third is expanding from pure shape acquisition toward intelligent perception, where geometry, optical properties, material characteristics, and functional information are jointly acquired and interpreted.
Looking ahead, the authors envision a future shaped by the synergy of hardware innovation and computational intelligence. With the development of high-speed projectors, advanced detectors, and computational optical components, 3D imaging systems are expected to evolve from data acquisition platforms into information encoding platforms through the co-optimization of hardware and algorithms.
Meanwhile, computational 3D imaging is expected to become a foundational framework for next-generation 3D vision. By combining physically interpretable computational imaging models with AI-powered inverse problem solvers, future systems will achieve both reliability and intelligence. Moving beyond geometry reconstruction, they will enable a deeper understanding of light transport, material properties, and scene formation mechanisms, opening new opportunities for intelligent perception technologies.
Light: Advanced Manufacturing
Review of Fringe Projection Profilometry: From geometric triangulation tocomputational 3D imaging