Ensemble of exemplar-SVMs for object detection and beyond

By Tomasz Malisiewicz, Abhinav Kumar Gupta, Alexei A. Efros. 2011

997 citations recorded by the source corpus.

This paper proposes a conceptually simple but surprisingly powerful method which combines the effectiveness of a discriminative object detector with the explicit correspondence offered by a nearest-neighbor approach. The method is based on training a separate linear SVM classifier for every exemplar in the training set. Each of these Exemplar-SVMs is thus defined by a single positive instance and millions of negatives. While each detector is quite specific to its exemplar, we empirically observe that an ensemble of such Exemplar-SVMs offers surprisingly good generalization. Our performance on the PASCAL VOC detection task is on par with the much more complex latent part-based model of Felzenszwalb et al., at only a modest computational cost increase. But the central benefit of our approach is that it creates an explicit association between each detection and a single training exemplar. Because most detections show good alignment to their associated exemplar, it is possible to transfer any available exemplar meta-data (segmentation, geometric structure, 3D model, etc.) directly onto the detections, which can then be used as part of overall scene understanding.

Subfields: Computer Vision and Pattern Recognition, Media Technology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Unsupervised Feature Learning via Non-parametric Instance Discrimination (2018, background): “We take the class-wise supervision to the extreme of instance-wise supervision, and ask: Can we learn a meaningful metric that reflects apparent similarity among instances via pure discriminative learning? An image is distinctive in its own right, and each could differ significantly from other image”
Training Region-Based Object Detectors with Online Hard Example Mining (2016, background): “The idea of dataset boot-strapping [25, 30], typically called hard negative mining in recent work [12], appears in the training of most successful object detectors [7, 9, 12, 13, 15, 16, 23, 25, 29].”
Dual Contradistinctive Generative Autoencoder (2020, neutral): “Inspired by the literature on instance-level classiﬁcation [50], approximating likelihood by classiﬁcation [60], and contrastive learning [20, 65, 22], we propose to model instance-level ﬁdelity by contrastive loss (commonly referred to as InfoNCE loss) [62].”
A critical analysis of self-supervision, or what we can learn from a single image (2019, methodology): “Single sample learning was pursued by the semi-parametric exemplar SVM model (Malisiewicz et al., 2011).”
Local Aggregation for Unsupervised Learning of Visual Embeddings (2019, background): “[55, 49, 14, 75]) do not benefit significantly from increasing depth.”
Learning feed-forward one-shot learners (2016, background): “These problems are usually tackled by using generative models [18, 13] or, in a discriminative setting, using ad-hoc solutions such as exemplar support vector machines (SVMs) [14].”
Deep Clustering for Unsupervised Learning of Visual Features (2018, methodology): “Their approach discriminates between images in a similar way as examplar SVM [36], while we are simply clustering them.”
Video Representation Learning Using Discriminative Pooling (2018, methodology): “In this respect, our pooling scheme is also different from ExemplarSVMs [29, 56, 61] that learns feature filters per data sample and then use these filters for feature extraction.”
HOGgles: Visualizing Object Detection Features (2013, background): “Consider the top detections for the exemplar object detector [9, 15] for a few images shown in Figure 5.”
DeepEdge: A multi-scale bifurcated deep network for top-down contour detection (2014, methodology): “This hold-out dataset is used for the hard-positive mining step [22].”
Computational Baby Learning (2014, methodology): “Inspired by [17], we train a separate linear SVM classifier for each positive instance, and each SVM classifier is highly tuned according to the exemplar’s appearance.”
Class-Balanced Loss Based on Effective Number of Samples (2019, background): “Another line of important work aims to study sample difﬁculty in terms of loss and assign higher weights to hard examples [11, 29, 8, 26].”
Improving Generalization via Scalable Neighborhood Component Analysis (2018, background): “Exemplar-based models have shown to be eﬀective for learning object classes [2] and object detection [25].”
Watch and learn: Semi-supervised learning of object detectors from videos (2015, methodology): “We start tracking from the conﬁdent ESVM detections to label new examples.”
High-for-Low and Low-for-High: Efficient Boundary Detection from Deep Object Features and Its Applications to High-Level Vision (2015, methodology): “We use this for the hard-positive mining [21] in order to reduce the number of false-negative predictions.”
EX2: Exploration with Exemplar Models for Deep Reinforcement Learning (2017, methodology): “We borrow the term “exemplar model” from Malisiewicz et al. (2011), which coined the term “ex-emplar SVM” to refer to a particular linear model trained to classify each instance against all others.”
Audio-Visual Instance Discrimination with Cross-Modal Agreement (2020, methodology): “The learned representations are optimized for ‘instance discrimination’ [17, 51, 86], i.”
CliqueCNN: Deep Unsupervised Exemplar Learning (2016, methodology): “However, large intra-class variability has recently spurred exemplar methods [15, 11], which split this problem into simpler sub-tasks.”