About

Pedro F Felzenszwalb is a Professor of Engineering at Brown University, specializing in Artificial Intelligence, Machine Learning, Algorithms and Theory, Computer Vision, and Data Science. He is associated with the Computer Science department and can be contacted via phone at 401-863-1531 or email at pedro_felzenszwalb@brown.edu. His research focuses on advancing understanding and applications within these areas, contributing to the development of intelligent systems and computational methods. As a faculty member, he is involved in teaching, mentoring, and research activities that support the university's mission in computer science and engineering.

Research topics

• Computer Science
• Artificial Intelligence
• Mathematical optimization
• Statistics
• Computer vision
• Algorithm
• Discrete mathematics
• Mathematics
• Combinatorics
• Mathematical analysis

Selected publications

• A Burst-mode Cryogenic Thermal Imager Readout IC with Spatial and Temporal Compression

  2025-05-25

  article

  This paper presents a high-speed global-shutter readout integrated circuit (ROIC), which is designed to capture bursts at up to 5 million frames per second from a 32<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">V</sup> ×32<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">H</sup> pixel infrared focal plane array (FPA) detector. The ROIC hosts a 1000-frame on-chip analog burst memory bank, and has additional hardware capabilities for on-chip spatial or temporal compression to extend the supported burst recording duration. The system is intended to operate inside a cryostat at liquid nitrogen temperatures, to support high speed thermal imaging in the mid-wavelength infrared (MWIR) band.

  PublisherDOI

• Projected Subgradient Ascent for Convex Maximization

  arXiv (Cornell University) · 2025-11-01

  preprintOpen access1st authorCorresponding

  We consider the problem of maximizing a convex function over a closed convex set in a real Hilbert space. For linear functions, we show that a single orthogonal projection suffices to obtain an approximate solution. For continuous convex functions over convex sets, we show that projected subgradient ascent converges to a first-order stationary point when using arbitrarily large step sizes. Taking the step sizes to infinity leads to a deterministic variant of the conditional gradient algorithm, and iterated linear optimization as a special case.

  PublisherOA PDFDOI

• Super-Resolution with Structured Motion

  2025-07-21

  articleOpen access

  We consider the limits of super-resolution using imaging constraints. Due to various theoretical and practical limitations, reconstruction-based methods have been largely restricted to small increases in resolution. In addition, motion-blur is usually seen as a nuisance that impedes super-resolution. We show that by using high-precision motion information, sparse image priors, and convex optimization, it is possible to increase resolution by large factors. A key operation in super-resolution is deconvolution with a box. In general, convolution with a box is not invertible. However, we obtain perfect reconstructions of sparse signals using convex optimization. We also show that motion blur can be helpful for super-resolution. We demonstrate that using pseudo-random motion it is possible to reconstruct a high-resolution target using a single low-resolution image. We present numerical experiments with simulated data and results with real data captured by a camera mounted on a computer-controlled stage.

  PublisherOA PDFDOI

• Automated brightfield layerwise evaluation in three-dimensional micropatterning via two-photon polymerization

  Optics Express · 2024-03-12 · 5 citations

  articleOpen access

  Two-photon polymerization (TPP) is an advanced 3D fabrication technique capable of creating features with submicron precision. A primary challenge in TPP lies in the facile and accurate characterization of fabrication quality, particularly for structures possessing complex internal features. In this study, we introduce an automated brightfield layerwise evaluation technique that enables a simple-to-implement approach for in situ monitoring and quality assessment of TPP-fabricated structures. Our approach relies on sequentially acquired brightfield images during the TPP writing process and using background subtraction and image processing to extract layered spatial features. We experimentally validate our method by printing a fibrous tissue scaffold and successfully achieve an overall system-adjusted fidelity of 87.5% in situ. Our method is readily adaptable in most TPP systems and can potentially facilitate high-quality TPP manufacturing of sophisticated microstructures.

  PublisherDOI

• Spectral Image Segmentation with Global Appearance Modeling

  SSRN Electronic Journal · 2023-01-01 · 1 citations

  preprintOpen accessSenior author
  PublisherDOI

• Convex combination belief propagation

  Applied Mathematics and Computation · 2022-10-17

  articleSenior author
  PublisherDOI

• Direct Estimation of Appearance Models for Segmentation

  SIAM Journal on Imaging Sciences · 2022-02-14

  preprintOpen access

  Image segmentation algorithms often depend on appearance models that characterize the distribution of pixel values in different image regions. We describe a new approach for estimating appearance models directly from an image, without explicit consideration of the pixels that make up each region. Our approach is based on novel algebraic expressions that relate local image statistics to the appearance of spatially coherent regions. We describe two algorithms that can use the aforementioned algebraic expressions to estimate appearance models directly from an image. The first algorithm solves a system of linear and quadratic equations using a least squares formulation. The second algorithm is a spectral method based on an eigenvector computation. We present experimental results that demonstrate the proposed methods work well in practice and lead to effective image segmentation algorithms.

  PublisherOA PDFDOI

• Object Detection

  Springer eBooks · 2021 · 60 citations

  • Computer Science
  • Artificial Intelligence
  • Computer Science
  PublisherDOI

• Iterated linear optimization

  Quarterly of Applied Mathematics · 2021-05-06 · 1 citations

  preprintOpen access1st authorCorresponding

  We introduce a fixed point iteration process built on optimization of a linear function over a compact domain. We prove the process always converges to a fixed point and explore the set of fixed points in various convex sets. In particular, we consider elliptopes and derive an algebraic characterization of their fixed points. We show that the attractive fixed points of an elliptope are exactly its vertices. Finally, we discuss how fixed point iteration can be used for rounding the solution of a semidefinite programming relaxation.

  PublisherOA PDFDOI

• Convex Combination Belief Propagation Algorithms

  arXiv (Cornell University) · 2021-05-26

  preprintOpen accessSenior author

  We present new message passing algorithms for performing inference with graphical models. Our methods are designed for the most difficult inference problems where loopy belief propagation and other heuristics fail to converge. Belief propagation is guaranteed to converge when the underlying graphical model is acyclic, but can fail to converge and is sensitive to initialization when the underlying graph has complex topology. This paper describes modifications to the standard belief propagation algorithms that lead to methods that converge to unique solutions on graphical models with arbitrary topology and potential functions.

  PublisherOA PDFDOI

Recent grants

• CAREER: Object Recognition with Hierarchical Models

  NSF · $167k · 2011–2014

• CAREER: Object Recognition with Hierarchical Models

  NSF · $450k · 2008–2012

• RI: Medium: Collaborative Research: Graph Cut Algorithms for Domain-specific Higher Order Priors

  NSF · $330k · 2012–2016

Frequent coauthors

• Ross Girshick

  22 shared

• Caroline J. Klivans

  16 shared

• Alice Paul

  Brown University

  11 shared

• David McAllester

  10 shared

• Jeová Farias Sales Rocha Neto

  Bowdoin College

  8 shared

• Sobhan Naderi Parizi

  Google (United States)

  7 shared

• Daniel P. Huttenlocher

  7 shared

• Yali Amit

  University of Chicago

  6 shared

Labs

• pff's homepagePI

Education

• Ph.D., Computer Science

  University of California, Berkeley

  1999

• M.S., Computer Science

  University of California, Berkeley

  1994

• B.S., Electrical Engineering and Computer Science

  University of California, Berkeley

  1992