• Publications
  • Influence
High-Speed Tracking with Kernelized Correlation Filters
TLDR
We propose an analytic model for datasets of thousands of translated patches. Expand
  • 3,185
  • 957
  • PDF
Fully-Convolutional Siamese Networks for Object Tracking
TLDR
The problem of arbitrary object tracking has traditionally been tackled by learning a model of the object’s appearance exclusively online, using as sole training data the video itself. Expand
  • 1,562
  • 435
  • PDF
Exploiting the Circulant Structure of Tracking-by-Detection with Kernels
TLDR
We derive closed-form solutions for training and detection with several types of kernels, including the popular Gaussian and polynomial kernels. Expand
  • 1,635
  • 307
  • PDF
End-to-End Representation Learning for Correlation Filter Based Tracking
TLDR
We interpret the Correlation Filter learner, which has a closed-form solution, as a differentiable layer in a deep neural network. Expand
  • 799
  • 188
  • PDF
The Sixth Visual Object Tracking VOT2018 Challenge Results
TLDR
The Visual Object Tracking challenge VOT2018 is the sixth annual tracker benchmarking activity organized by the VOT initiative. Expand
  • 249
  • 71
  • PDF
The Visual Object Tracking VOT2017 Challenge Results
TLDR
The Visual Object Tracking challenge VOT2017 is the fifth annual tracker benchmarking activity organized by the VOT initiative. Expand
  • 286
  • 66
  • PDF
Meta-learning with differentiable closed-form solvers
TLDR
In this paper, we propose to use these fast convergent methods as the main adaptation mechanism for few-shot learning. Expand
  • 211
  • 65
  • PDF
Invariant Information Clustering for Unsupervised Image Classification and Segmentation
TLDR
We present a novel clustering objective that learns a neural network classifier from scratch, given only unlabelled data samples. Expand
  • 140
  • 31
  • PDF
Learning feed-forward one-shot learners
TLDR
In this paper, we propose a method to learn the parameters of a deep model from a single exemplar in a learning to learn formulation. Expand
  • 261
  • 20
  • PDF
Long-term Tracking in the Wild: A Benchmark
TLDR
We introduce the OxUvA dataset and benchmark for evaluating single-object tracking algorithms, considering both the ability to locate the target and to determine whether it is present or absent. Expand
  • 57
  • 17
  • PDF