Spectrum from Defocus: Fast Spectral Imaging with Chromatic Focal Stack

CVPR, 2026
1Northwestern University, 2Dolby Laboratories, Inc., 3Purdue University *Equal contribution

SfD achieves state-of-the-art hyperspectral imaging with two off-the-shelf lenses, a grayscale sensor, and under one second reconstruction.

Abstract

Hyperspectral cameras face harsh trade-offs between spatial, spectral, and temporal resolution in inherently low-photon conditions. Computational imaging systems break through these trade-offs with compressive sensing, but have required complex optics and/or extensive compute.

We present Spectrum from Defocus (SfD), a chromatic focal sweep method that achieves state-of-the-art hyperspectral imaging with only two off-the-shelf lenses, a grayscale sensor, and less than one second of reconstruction time. By capturing a chromatically-aberrated focal stack that preserves nearly all incident light, and reconstructing it with a fast physics-based iterative algorithm, SfD delivers sharp, accurate hyperspectral images.

The combination of photon efficiency, optical simplicity, and physical interpretability makes SfD a promising solution for fast, compact, interpretable hyperspectral imaging.

Optical Setup

Drag the slider to move the second lens. Different wavelengths come into focus on the grayscale sensor, creating wavelength-dependent blur that serves as our optical encoding.

λ in focus: — nm

Chromatic focal shift and lens motion are exaggerated for clarity compared to the paper.

Method

Coming soon.

Example Results

Hover over the reconstruction to see ground truth and reconstructed spectra at that pixel. This may take a while to load, hyperspectral images are large files to download.

Spectral lines are available on the desktop interface.

Loading results...

Measurements

Reconstruction (RGB)

PSNR: 27.15 dB

Spectrum

gt recon · Hover over image →

Measurements

Reconstruction (RGB)

PSNR: 33.57 dB

Spectrum

gt recon · Hover over image →

Poster

Coming soon.