• Publications
  • Influence
Performance of optical flow techniques
TLDR
We compare a number of commonly cited optical flow techniques, including instances of differential, matching, energy-based, and phase-based methods. Expand
  • 4,003
  • 238
  • PDF
TurboPixels: Fast Superpixels Using Geometric Flows
TLDR
We describe a geometric-flow-based algorithm for computing a dense oversegmentation of an image, often referred to as superpixels. Expand
  • 1,004
  • 128
  • PDF
cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination
TLDR
We show 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. Expand
  • 1,349
  • 100
  • PDF
Minimal Loss Hashing for Compact Binary Codes
TLDR
We propose a method for learning similarity-preserving hash functions that map high-dimensional data onto binary codes. Expand
  • 688
  • 93
  • PDF
Gaussian Process Dynamical Models for Human Motion
TLDR
We introduce Gaussian process dynamical models (GPDMs) for nonlinear time series analysis, with applications to learning models of human pose and motion from high-dimensional motion capture data. Expand
  • 840
  • 91
  • PDF
VSE++: Improving Visual-Semantic Embeddings with Hard Negatives
TLDR
We present a new technique for learning visual-semantic embeddings for cross-modal retrieval. Expand
  • 304
  • 78
  • PDF
Gaussian Process Dynamical Models
TLDR
This paper introduces Gaussian Process Dynamical Models (GPDM) for nonlinear time series analysis. Expand
  • 463
  • 60
  • PDF
Cartesian K-Means
TLDR
A fundamental limitation of quantization techniques like the k-means clustering algorithm is the storage and run-time cost associated with large numbers of clusters required to keep quantization errors small and model fidelity high. Expand
  • 282
  • 59
  • PDF
Robust Online Appearance Models for Visual Tracking
TLDR
We propose a framework for learning robust, adaptive, appearance models to be used for motion-based tracking of natural objects. Expand
  • 1,160
  • 56
  • PDF
VSE++: Improved Visual-Semantic Embeddings
TLDR
This paper investigates the problem of image-caption retrieval using joint visualsemantic embeddings. Expand
  • 139
  • 49
  • PDF
...
1
2
3
4
5
...