OpenGAN: Open-Set Recognition via Open Data Generation

@article{Kong2021OpenGANOR,
  title={OpenGAN: Open-Set Recognition via Open Data Generation},
  author={Shu Kong and Deva Ramanan},
  journal={2021 IEEE/CVF International Conference on Computer Vision (ICCV)},
  year={2021},
  pages={793-802}
}
  • Shu Kong, D. Ramanan
  • Published 7 April 2021
  • Computer Science
  • 2021 IEEE/CVF International Conference on Computer Vision (ICCV)
Real-world machine learning systems need to analyze novel testing data that differs from the training data. In K-way classification, this is crisply formulated as open-set recognition, core to which is the ability to discriminate open-set data outside the K closed-set classes. Two conceptually elegant ideas for open-set discrimination are: 1) discriminatively learning an open-vs-closed binary discriminator by exploiting some outlier data as the open-set, and 2) unsupervised learning the closed… 
Open-world Machine Learning: Applications, Challenges, and Opportunities
TLDR
This paper will help researchers to understand the comprehensive developments of open-world learning and likelihoods to extend the research in suitable areas and will also help to select applicable methodologies and datasets to explore this further.
Generalized Category Discovery
TLDR
This paper establishes strong baselines by taking state-of-the-art algorithms from novel category discovery and adapting them for this task, and introduces a simple yet effective semi-supervised k-means method to cluster the unlabelled data into seen and unseen classes automatically, substantially outperforming the baselines.
Long-Tailed Recognition via Weight Balancing
TLDR
An orthogonal direction, weight balancing, is explored by the empirical observation that the naively trained classifier has “artificially” larger weights in norm for common classes (because there exists abundant data to train them, unlike the rare classes) and achieves the state-ofthe-art accuracy on five standard benchmarks.
OpenTAL: Towards Open Set Temporal Action Localization
TLDR
This paper steps toward the Open Set TAL (OSTAL) problem and proposes a general framework OpenTAL based on Evidential Deep Learning (EDL), which consists of uncertainty-aware action classification, actionness prediction, and temporal location regression.
FLUID: A Unified Evaluation Framework for Flexible Sequential Data
TLDR
A new unified evaluation framework – FLUID (Flexible Sequential Data), which integrates the objectives of few-shot, continual, transfer, and representation learning while enabling comparison and integration of techniques across these subfields and presents two new baselines which outperform other evaluated methods on FLUID.
A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection: Solutions and Future Challenges
TLDR
This survey aims to provide a cross-domain and comprehensive review of numerous eminent works in respective areas while identifying their commonalities and discusses and shed light on future lines of research, intending to bring these fields closer together.
Generalized Out-of-Distribution Detection: A Survey
TLDR
This survey presents a generic framework called generalized OOD detection, which encompasses the five aforementioned problems, i.e., AD, ND, OSR, OOD Detection, and OD, and conducts a thorough review of each of the five areas by summarizing their recent technical developments.
GOSS: Towards Generalized Open-set Semantic Segmentation
TLDR
This paper proposes GOSS, which unifies the abilities of two well-defined segmentation tasks, i.e., OSS and generic segmentation (GS), in a holistic way, and proposes a metric which balances the pixel classification and clustering aspects.
Expanding Low-Density Latent Regions for Open-Set Object Detection
TLDR
This work proposes to identify unknown objects by separating high/low-density regions in the latent space, based on the consensus that unknown objects are usually distributed in low-density latent regions, and presents a novel Open-set Detector (OpenDet) with expanded low- density regions.
Open-World Instance Segmentation: Exploiting Pseudo Ground Truth From Learned Pairwise Affinity
TLDR
This work proposes a novel approach for mask proposals, Generic Grouping Networks (GGNs), con-structed without semantic supervision, which combines a local measure of pixel affinity with instance-level mask supervision, producing a training regimen designed to make the model as generic as the data diversity allows.
...
1
2
...

References

SHOWING 1-10 OF 73 REFERENCES
Generative-Discriminative Feature Representations for Open-Set Recognition
TLDR
This work proposes two techniques to force class activations of open-set samples to be low, and uses self-supervision to force the network to learn more informative featues when assigning class scores to improve separation of classes from each other and from open- set samples.
Classification-Reconstruction Learning for Open-Set Recognition
TLDR
This work utilizes latent representations for reconstruction and enables robust unknown detection without harming the known-class classification accuracy, and outperforms existing deep open-set classifiers in multiple standard datasets and is robust to diverse outliers.
An Empirical Exploration of Open-Set Recognition via Lightweight Statistical Pipelines
TLDR
An empirical exploration of open-set classification is carried out, and it is found that combining classic statistical methods with carefully computed features can dramatically outperform prior work.
Recent Advances in Open Set Recognition: A Survey
TLDR
This paper provides a comprehensive survey of existing open set recognition techniques covering various aspects ranging from related definitions, representations of models, datasets, evaluation criteria, and algorithm comparisons to highlight the limitations of existing approaches and point out some promising subsequent research directions.
Generative OpenMax for Multi-Class Open Set Classification
TLDR
The proposed method, called G-OpenMax, extends OpenMax by employing generative adversarial networks (GANs) for novel category image synthesis and provides a way to visualize samples representing the unknown classes from open space.
Toward Open Set Recognition
TLDR
This paper explores the nature of open set recognition and formalizes its definition as a constrained minimization problem, and introduces a novel “1-vs-set machine,” which sculpts a decision space from the marginal distances of a 1-class or binary SVM with a linear kernel.
Learning Open Set Network with Discriminative Reciprocal Points
TLDR
A new concept, Reciprocal Point, which is the potential representation of the extra-class space corresponding to each known category, which can indirectly introduce the unknown information into the learner with only known classes, so as to learn more compact and discriminative representations.
C2AE: Class Conditioned Auto-Encoder for Open-Set Recognition
TLDR
An open-set recognition algorithm using class conditioned auto-encoders with novel training and testing methodologies is proposed and experiments show that the proposed method performs significantly better than the state of the art methods.
Towards Open World Recognition
  • Abhijit Bendale, T. Boult
  • Computer Science
    2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2015
TLDR
It is proved that thresholding sums of monotonically decreasing functions of distances in linearly transformed feature space can balance “open space risk” and empirical risk and it is presented the Nearest Non-Outlier (NNO) algorithm that evolves model efficiently, adding object categories incrementally while detecting outliers and managing open space risk.
Towards Open Set Deep Networks
  • Abhijit Bendale, T. Boult
  • Computer Science
    2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2016
TLDR
The proposed OpenMax model significantly outperforms open set recognition accuracy of basic deep networks as well as deep networks with thresholding of SoftMax probabilities, and it is proved that the OpenMax concept provides bounded open space risk, thereby formally providing anopen set recognition solution.
...
1
2
3
4
5
...