Leveraging computational imaging, SLIM enables 3D snapshot transient imaging with full light-field capabilities—including depth reconstruction, post-capture refocusing, and extended depth of field. By replacing the conventional camera in an imaging system, SLIM adds high-speed 3D acquisition to a wide range of modalities. The development of SLIM is motivated by the need for faster light-field imaging. Traditional light-field cameras capture the full four-dimensional plenoptic function (two spatial and two angular dimensions), requiring large-format sensors and resulting in bulky datasets and low frame rates.

​

SLIM addresses this data bottleneck by recognizing a key fact that the multiple perspective views in a conventional light-field camera are highly redundant, differing mainly in depth-related disparities. SLIM reduces this redundancy by compressing perspective images along various orientations and encoding them into distinct views. This drastically reduces the data volume, enabling high-resolution 3D imaging using a low-pixel-count camera with high readout speed.

​

A typical SLIM system uses an array of dove prisms and lenslets to form rotated perspective images. Each dove prism is oriented at a unique angle, generating rotated views that converge at an intermediate image plane. These views are then passed through an anamorphic relay, which compresses the image array along one axis while maintaining scale along the orthogonal direction. This rescaled image is recorded by a 2D sensor with a much smaller pixel count than that required for direct measurement of the full image array.

​