One-shot learning of object categories

  title={One-shot learning of object categories},
  author={Li Fei-Fei and Rob Fergus and Pietro Perona},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
Learning visual models of object categories notoriously requires hundreds or thousands of training examples. We show that it is possible to learn much information about a category from just one, or a handful, of images. The key insight is that, rather than learning from scratch, one can take advantage of knowledge coming from previously learned categories, no matter how different these categories might be. We explore a Bayesian implementation of this idea. Object categories are represented by… 

Attribute-Based Transfer Learning for Object Categorization with Zero/One Training Example

This paper builds a generative attribute model to learn the probabilistic distributions of image features for each attribute, which can be used to classify unseen images of target categories or facilitate learning classifiers for target categories when there is only one training examples per target category (one-shot learning).

One-Shot Concept Learning by Simulating Evolutionary Instinct Development

A computational model is proposed that can successfully learn an object category from as few as one example and allows its learning style to be tailored explicitly to a scenario and it can be applied to essentially any class of visual objects.

One-Shot Learning with a Hierarchical Nonparametric Bayesian Model

We develop a hierarchical Bayesian model that learns categories from single training examples. The model transfers acquired knowledge from previously learned categories to a novel category, in the

One Shot Learning via Compositions of Meaningful Patches

  • A. WongA. Yuille
  • Computer Science
    2015 IEEE International Conference on Computer Vision (ICCV)
  • 2015
This work proposes an unsupervised method for learning a compact dictionary of image patches representing meaningful components of an objects and builds a compositional model that outperforms a number of popular algorithms on a one-shot learning task.

Improved Object Categorization and Detection Using Comparative Object Similarity

It is shown that local object similarity information--statements that pairs of categories are similar or dissimilar--is a very useful cue to tie different categories to each other for effective knowledge transfer.

Category Modeling from Just a Single Labeling: Use Depth Information to Guide the Learning of 2D Models

A new strategy that uses the depth information in RGBD images to guide the model learning for object detection in ordinary RGB images is proposed and achieves superior performance as good as the supervised methods that require the labeling of all target objects.

Gaussian Processes for Object Categorization

This work shows that with an appropriate combination of kernels a significant boost in classification performance is possible, and indicates the utility of active learning with probabilistic predictive models, especially when the amount of training data labels that may be sought for a category is ultimately very small.

Optimizing one-shot recognition with micro-set learning

This work details a discriminative approach for optimizing one-shot recognition using micro-sets and presents experiments on the Animals with Attributes and Caltech-101 datasets that demonstrate the benefits of the formulation.

Learning Categories From Few Examples With Multi Model Knowledge Transfer

This paper presents a discriminative model adaptation algorithm able to proficiently learn a target object with few examples by relying on other previously learned source categories by solving a convex optimization problem.



A Bayesian approach to unsupervised one-shot learning of object categories

This work presents a method for learning object categories from just a few images, based on incorporating "generic" knowledge which may be obtained from previously learnt models of unrelated categories, in a variational Bayesian framework.

Learning Generative Visual Models from Few Training Examples: An Incremental Bayesian Approach Tested on 101 Object Categories

Unsupervised learning of models for object recognition

A method to learn object class models from unlabeled and unsegmented cluttered cluttered scenes for the purpose of visual object recognition achieves very good classification results on human faces, cars, leaves, handwritten letters, and cartoon characters.

Unsupervised Learning of Models for Recognition

A method to learn object class models from unlabeled and unsegmented cluttered cluttered scenes for the purpose of visual object recognition that achieves very good classification results on human faces and rear views of cars.

A sparse object category model for efficient learning and exhaustive recognition

We present a "parts and structure" model for object category recognition that can be learnt efficiently and in a semi-supervised manner: the model is learnt from example images containing category

Object class recognition by unsupervised scale-invariant learning

The flexible nature of the model is demonstrated by excellent results over a range of datasets including geometrically constrained classes (e.g. faces, cars) and flexible objects (such as animals).

A Computational Model for Visual Selection

The model was not conceived to explain brain functions, but it does cohere with evidence about the functions of neurons in V1 and V2, such as responses to coarse or incomplete patterns and to scale and translation invariance in IT.

Viewpoint-invariant learning and detection of human heads

A method to learn models of human heads for the purpose of detection from different viewing angles using a model where objects are represented as constellations of rigid features (parts) using a joint probability density function on the shape of the constellation.

Rapid object detection using a boosted cascade of simple features

  • Paul A. ViolaMichael J. Jones
  • Computer Science
    Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001
  • 2001
A machine learning approach for visual object detection which is capable of processing images extremely rapidly and achieving high detection rates and the introduction of a new image representation called the "integral image" which allows the features used by the detector to be computed very quickly.

Sharing features: efficient boosting procedures for multiclass object detection

  • A. TorralbaK. MurphyW. Freeman
  • Computer Science
    Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004.
  • 2004
A multi-class boosting procedure (joint boosting) is presented that reduces both the computational and sample complexity, by finding common features that can be shared across the classes.