Author pages are created from data sourced from our academic publisher partnerships and public sources.
Share This Author
Performance of optical flow techniques
- J. Barron, David J. Fleet, S. Beauchemin
- MathematicsInternational Journal of Computer Vision
- 1 February 1994
These comparisons are primarily empirical, and concentrate on the accuracy, reliability, and density of the velocity measurements; they show that performance can differ significantly among the techniques the authors implemented.
cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination
- A. Punjani, J. Rubinstein, David J. Fleet, Marcus A. Brubaker
- Computer ScienceNature Methods
- 6 February 2017
It is shown that stochastic gradient descent (SGD) and branch-and-bound maximum likelihood optimization algorithms permit the major steps in cryo-EM structure determination to be performed in hours or minutes on an inexpensive desktop computer.
VSE++: Improving Visual-Semantic Embeddings with Hard Negatives
A simple change to common loss functions used for multi-modal embeddings, inspired by hard negative mining, the use of hard negatives in structured prediction, and ranking loss functions, is introduced, which yields significant gains in retrieval performance.
TurboPixels: Fast Superpixels Using Geometric Flows
- Alex Levinshtein, Adrian Stere, Kiriakos N. Kutulakos, David J. Fleet, Sven J. Dickinson, Kaleem Siddiqi
- MathematicsIEEE Transactions on Pattern Analysis and Machine…
- 1 December 2009
A geometric-flow-based algorithm for computing a dense oversegmentation of an image, often referred to as superpixels, which yields less undersegmentation than algorithms that lack a compactness constraint while offering a significant speedup over N-cuts, which does enforce compactness.
Gaussian Process Dynamical Models for Human Motion
- Jack M. Wang, David J. Fleet, Aaron Hertzmann
- Computer ScienceIEEE Transactions on Pattern Analysis and Machine…
- 1 February 2008
This work marginalize out the model parameters in closed form by using Gaussian process priors for both the dynamical and the observation mappings, which results in a nonparametric model for dynamical systems that accounts for uncertainty in the model.
Minimal Loss Hashing for Compact Binary Codes
The formulation is based on structured prediction with latent variables and a hinge-like loss function that is efficient to train for large datasets, scales well to large code lengths, and outperforms state-of-the-art methods.
Gaussian Process Dynamical Models
This paper marginalize out the model parameters in closed-form, using Gaussian Process (GP) priors for both the dynamics and the observation mappings, resulting in a nonparametric model for dynamical systems that accounts for uncertainty in the model.
- Mohammad Norouzi, David J. Fleet
- Computer ScienceIEEE Conference on Computer Vision and Pattern…
- 23 June 2013
New models with a compositional parameterization of cluster centers are developed, so representational capacity increases super-linearly in the number of parameters, allowing one to effectively quantize data using billions or trillions of centers.
Robust Online Appearance Models for Visual Tracking
- A. Jepson, David J. Fleet, Thomas F. El-Maraghi
- Computer ScienceIEEE Trans. Pattern Anal. Mach. Intell.
- 1 October 2003
A framework for learning robust, adaptive, appearance models to be used for motion-based tracking of natural objects to provide robustness in the face of image outliers, while adapting to natural changes in appearance such as those due to facial expressions or variations in 3D pose.
VSE++: Improved Visual-Semantic Embeddings
This paper introduces a very simple change to the loss function used in the original formulation by Kiros et al. (2014), which leads to drastic improvements in the retrieval performance, and shows that similar improvements also apply to the Order-embeddings by Vendrov etAl.