I research how light interacts with materials and surfaces, using optical techniques to study their properties for scientific and technological applications.

This project studies how light reflects inside a right-angle prism to analyze materials, using the principle of total internal reflection to measure how samples absorb and transmit light.

Material Classification

This research uses time-resolved imaging to identify and separate materials in a scene. 

By capturing a transient histogram—a record of when photons arrive—we can analyze how different materials reflect light over time. A 1-D convolutional neural network processes this data to classify materials with high accuracy, even when they appear similar in color or texture.

Transient Imaging

This transient imaging setup demonstrates how light propagates through layered materials by capturing the time-resolved reflection of light at picosecond resolution. In this scene, the first layer is a transparent object placed in front of a second layer (an apple). Using a SPAD (Single-Photon Avalanche Diode) sensor and pulsed laser, the system records how photons reflect and scatter as they pass through the transparent layer and subsequently reach the apple behind it. The right-side image visualizes this light propagation, showing the interaction with the transparent layer first, followed by delayed reflections from the apple. 

NLOS

This experiment demonstrates a Non-Line-of-Sight (NLOS) imaging setup using a SPAD (Single-Photon Avalanche Diode) sensor and a pulsed laser to detect hidden objects. The laser emits pulses onto a visible relay surface (the wall), from which light scatters to reach a hidden target behind an occluder. The SPAD then captures the time-resolved return signal after the light reflects back from the target via the wall. On the right, the heatmap shows the accumulated photon counts over space, where brighter regions correspond to early and strong reflections