top of page

Temporal Imaging & Hidden Geometry

Capturing light transport at picosecond scales and reconstructing structure from indirect illumination — where time becomes a controllable dimension of imaging.

Femophotography_edited.jpg
Time  as
reconstruction

Conventional imaging integrates photons over time, collapsing dynamic propagation into a single frame. By resolving light transport at picosecond scales, time becomes a controllable dimension of capture rather than an invisible assumption. Imaging shifts from recording appearance to reconstructing events as they unfold in space–time.

Light transport   as
measurable structure

Light is typically treated as instantaneous illumination. When propagation is resolved, scattering, reflection, and interreflection emerge as structured signals with temporal order. Transport ceases to be an approximation embedded in shading models — it becomes measurable geometry encoded in delay.

Indirect visibility  as
inference

Once the speed of light is observable, indirect illumination becomes computationally interpretable. Geometry and material interactions can be inferred from propagation paths rather than direct line-of-sight capture. Visibility expands from what is seen directly to what can be reconstructed from transient signal structure.

These principles extend beyond imaging — they establish a framework for composing signal, motion, and visibility as controllable dimensions within experience systems.

Femto-Photography

Resolved light propagation at picosecond scales and introduced computational reconstruction of space–time volumes, establishing transient imaging as a measurable capture modality.

Winner of SIGGRAPH 2024 Test of Time award

Screenshot 2026-02-16 at 6.46.40 PM.png
Imaging the Propagation of Light Through Scenes at Picosecond Resolution

Articulated transient imaging as a systems framework, detailing capture architecture, time-unwarping, and reconstruction methods for physically consistent light transport analysis.

Screenshot 2026-02-16 at 6.48.54 PM.png
Estimating Motion and Size of Moving Non-Line-of-Sight Objects in Cluttered Environments

Demonstrated inference of hidden geometry and motion from indirect temporal signals, extending imaging beyond line-of-sight constraints through structured light transport analysis.

siggraphtalk_splash_removebg.png
This work established a discipline of time-based capture and reconstruction that treats motion and perception as composable signal — not just visual output.
bottom of page