AKAmit Kohli

Amit Kohli

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Amit Kohli has a total of 59 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Chromatix: a differentiable, GPU-accelerated wave-optics library

    by Diptodip Deb, Gert-Jan Both, Eric Bezzam, Amit Kohli, Siqi Yang, Amey Chaware, Cédric Allier, Changjia Cai, Geneva Anderberg, M. Hossein Eybposh, et al.

    Many current microscopy methods incorporate computational modeling as an integral part of the imaging process, either to solve inverse problems or optimize the optical system design itself. These methods often depend on differentiable optics simulations, yet no standardized framework exists, forcing computational optics researchers to repeatedly and independently implement simulations with limited reusability and performance. These common problems limit the potential impact of computational optics as a field. Here we present Chromatix: an open-source, graphics processing unit (GPU)-accelerated, differentiable wave-optics simulation library. Chromatix builds on JAX to democratize fast, parallelized simulation of diverse optical systems and expand the design space in computational optics. Chromatix standardizes a growing collection of optical elements and propagation methods allowing a broad range of applications, which we demonstrate here for snapshot microscopy, holography and phase retrieval. We demonstrate speed improvements of 2–6 times on a single GPU and up to 22 times on 8 GPUs.


  • Open Access

    Chromatix: a differentiable, GPU-accelerated wave-optics library

    by Diptodip Deb, Gert-Jan Both, Eric Bezzam, Amit Kohli, Siqi Yang, Amey Chaware, Cédric Allier, Changjia Cai, Geneva Anderberg, M. Hossein Eybposh, et al.

    Modern microscopy methods incorporate computational modeling as an integral part of the imaging process, either to solve inverse problems or optimize the optical system design itself. These methods often depend on differentiable optics simulations, yet no standardized framework exists—forcing computational optics researchers to repeatedly and independently implement simulations with limited reusability and performance. These common problems limit the potential impact of computational optics as a field. Here we present Chromatix: an open-source, GPU-accelerated, differentiable wave optics simulation library. Chromatix builds on JAX to democratize fast, parallelized simulation of diverse optical systems and expand the design space in computational optics. Chromatix standardizes a growing collection of optical elements and propagation methods allowing a broad range of applications, which we demonstrate here for snapshot microscopy, holography, and phase retrieval. We demonstrate speed improvements of 2-6× on a single GPU and up to 22× on 8 GPUs.


  • Top Cited Papers
    Open Access

    Ring deconvolution microscopy: exploiting symmetry for efficient spatially varying aberration correction

    by Amit Kohli, Anastasios N. Angelopoulos, David McAllister, Esther Whang, Sixian You, Kyrollos Yanny, Federico M. Gasparoli, Bo-Jui Chang, Reto Fiolka, Laura Waller

    The most ubiquitous form of aberration correction for microscopy is deconvolution; however, deconvolution relies on the assumption that the system’s point spread function is the same across the entire field of view. This assumption is often inadequate, but space-variant deblurring techniques generally require impractical amounts of calibration and computation. We present an imaging pipeline that leverages symmetry to provide simple and fast spatially varying deblurring. Our ring deconvolution microscopy method utilizes the rotational symmetry of most microscopes and cameras, and naturally extends to sheet deconvolution in the case of lateral symmetry. We derive theory and algorithms for ring deconvolution microscopy and propose a neural network based on Seidel aberration coefficients as a fast alternative. We demonstrate improvements in speed and image quality as compared to standard deconvolution and existing spatially varying deblurring across a diverse range of microscope modalities, including miniature microscopy, multicolor fluorescence microscopy, multimode fiber micro-endoscopy and light-sheet fluorescence microscopy. Our approach enables near-isotropic, subcellular resolution in each of these applications.


  • Top Cited Papers

    Noise2Image: noise-enabled static scene recovery for event cameras

    by Ruiming Cao, Dekel Galor, Amit Kohli, Jacob L. Yates, Laura Waller

    Event cameras, also known as dynamic vision sensors, are an emerging modality for measuring fast dynamics asynchronously. Event cameras capture changes of log-intensity over time as a stream of “events” and generally cannot measure intensity itself; hence, they are only used for imaging dynamic scenes. However, fluctuations due to random photon arrival inevitably trigger noise events, even for static scenes. While previous efforts have been focused on filtering out these undesirable noise events to improve signal quality, we find that, in the photon-noise regime, these noise events are correlated with the static scene intensity. We analyze the noise event generation and model its relationship to illuminance. Based on this understanding, we propose a method, called Noise2Image, to leverage the illuminance-dependent noise characteristics to recover the static parts of a scene, which are otherwise invisible to event cameras. We experimentally collect a dataset of noise events on static scenes to train and validate Noise2Image. Our results show that Noise2Image can robustly recover intensity images solely from noise events, providing an approach for capturing static scenes in event cameras, without additional hardware.


  • Open Access

    Event Based, Near Eye Gaze Tracking Beyond 10,000Hz

    by Anastasios N. Angelopoulos, Julien N. P. Martel, Amit P. S. Kohli, Jorg Conradt, Gordon Wetzstein

    The cameras in modern gaze-tracking systems suffer from fundamental bandwidth and power limitations, constraining data acquisition speed to 300 Hz realistically. This obstructs the use of mobile eye trackers to perform, e.g., low latency predictive rendering, or to study quick and subtle eye motions like microsaccades using head-mounted devices in the wild. Here, we propose a hybrid frame-event-based near-eye gaze tracking system offering update rates beyond 10,000 Hz with an accuracy that matches that of high-end desktop-mounted commercial trackers when evaluated in the same conditions. Our system builds on emerging event cameras that simultaneously acquire regularly sampled frames and adaptively sampled events. We develop an online 2D pupil fitting method that updates a parametric model every one or few events. Moreover, we propose a polynomial regressor for estimating the point of gaze from the parametric pupil model in real time. Using the first event-based gaze dataset, available at https://github.com/aangelopoulos/event_based_gaze_tracking , we demonstrate that our system achieves accuracies of 0.45 degrees--1.75 degrees for fields of view from 45 degrees to 98 degrees. With this technology, we hope to enable a new generation of ultra-low-latency gaze-contingent rendering and display techniques for virtual and augmented reality.

  • Open Access

    Semantic Implicit Neural Scene Representations With Semi-Supervised Training

    by Amit Kohli, Vincent Sitzmann, Gordon Wetzstein

    The recent success of implicit neural scene representations has presented a viable new method for how we capture and store 3D scenes. Unlike conventional 3D representations, such as point clouds, which explicitly store scene properties in discrete, localized units, these implicit representations encode a scene in the weights of a neural network which can be queried at any coordinate to produce these same scene properties. Thus far, implicit representations have primarily been optimized to estimate only the appearance and/or 3D geometry information in a scene. We take the next step and demonstrate that an existing implicit representation (SRNs) is actually multi-modal; it can be further leveraged to perform per-point semantic segmentation while retaining its ability to represent appearance and geometry. To achieve this multi-modal behavior, we utilize a semi-supervised learning strategy atop the existing pre-trained scene representation. Our method is simple, general, and only requires a few tens of labeled 2D segmentation masks in order to achieve dense 3D semantic segmentation. We explore two novel applications for this semantically aware implicit neural scene representation: 3D novel view and semantic label synthesis given only a single input RGB image or 2D label mask, as well as 3D interpolation of appearance and semantics.

  • Shift-Variant Deblurring for Rotationally Symmetric Systems

    by Amit Kohli, Anastasios Angelopoulos, Sixian You, Laura Waller

    We present a fast image deblurring method for rotationally symmetric systems with spatially-varying aberrations. We calibrate our method with point spread function measurements along a line. Our method outperforms standard deconvolution on the UCLA Miniscope.