Detection Evolution with Multi-order Contextual Co-occurrence

Chen, Guang; Ding, Yuanyuan; Xiao, Jing; Han, Tony Xiao

doi:10.1109/cvpr.2013.235

Cited by 78 publications

(52 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[5] propose a technique for exploiting contextual information from a single-category detector output. Although the method itself does not directly use inter-class information, they utilize the detections given by the contextual rescoring approach of [16].…”

Section: Evaluation and Comparison To Existing Workmentioning

confidence: 99%

Segmentation Driven Object Detection with Fisher Vectors

Cinbiş

Verbeek

Schmid

2013

2013 IEEE International Conference on Computer Vision

102

View full text Add to dashboard Cite

We present an object detection system based on the Fisher vector (FV) image representation computed over SIFT and color descriptors. For computational and storage efficiency, we use a recent segmentation-based method to generate class-independent object detection hypotheses, in combination with data compression techniques. Our main contribution is a method to produce tentative object segmentation masks to suppress background clutter in the features. Re-weighting the local image features based on these masks is shown to improve object detection significantly. We also exploit contextual features in the form of a full-image FV descriptor, and an inter-category rescoring mechanism. Our experiments on the PASCAL VOC 2007 and 2010 datasets show that our detector improves over the current state-of-the-art detection results.

show abstract

Section: Evaluation and Comparison To Existing Workmentioning

confidence: 99%

Segmentation Driven Object Detection with Fisher Vectors

Cinbiş

Verbeek

Schmid

2013

2013 IEEE International Conference on Computer Vision

102

View full text Add to dashboard Cite

show abstract

“…We evaluate results using the standard PASCAL scheme based on bounding box overlap that produces precision-recall curves and Average Precision (AP) measures, as done by [11] and [10]. State-of-the-art results on this dataset involve improved features (irregular HOG grids, additional channels, etc) [5,20] and use non-linear classifiers [5,20], deformable parts [11], or context [7]. However, we focus on rigid templates of HOG features on a regular grid [11], and linear SVM for two reasons: 1) to isolate the contribution of CDF (as opposed to additional machinery such as deformable parts, context, exemplars), 2) simple HOG features and linear SVM are still prevalent as building blocks in state-of-theart approaches (as is evident in section 1.1).…”

Section: Methodsmentioning

confidence: 99%

Composite Discriminant Factor analysis

Morariu

Ahmed

Santhanam

et al. 2014

IEEE Winter Conference on Applications of Computer Vision

View full text Add to dashboard Cite

We propose a linear dimensionality reduction method, Composite Discriminant Factor (CDF) analysis, which searches for a discriminative but compact feature subspace that can be used as input to classifiers that suffer from problems such as multi-collinearity or the curse of dimensionality. The subspace selected by CDF maximizes the performance of the entire classification pipeline, and is chosen from a set of candidate subspaces that are each discriminative. Our method is based on Partial Least Squares (PLS) analysis, and can be viewed as a generalization of the PLS1 algorithm, designed to increase discrimination in classification tasks. We demonstrate our approach on the UCF50 action recognition dataset, two object detection datasets (INRIA pedestrians and vehicles from aerial imagery), and machine learning datasets from the UCI Machine Learning repository. Experimental results show that the proposed approach improves significantly in terms of accuracy over linear SVM, and also over PLS in terms of compactness and efficiency, while maintaining or improving accuracy.

show abstract

“…We compare with the best-performing methods as suggested by the Caltech and ETH benchmarks 2 , which report the top results of these two datasets, including VJ [34], HOG [5], DBN-Isol [24], ACF [6], ACF-Caltech [6], MultiFtr+CSS [35], MultiResC [28], Roerei [1], MOCO [3], MT-DPM [38], ChnFtrs [8], HogLbp [36], Pls [29], CrossTalk [7], LatSVM-V2 [11], MLS [22], ConvNet [30], and UDN [25]. All of these approaches detect pedestrians on static images, like our method, rather than using video motion as additional information.…”

Section: Methodsmentioning

confidence: 99%

Switchable Deep Network for Pedestrian Detection

Luo

Tian

Wang

et al. 2014

2014 IEEE Conference on Computer Vision and Pattern Recognition

210

123

View full text Add to dashboard Cite

In this paper, we propose a Switchable Deep Network (SDN) for pedestrian detection. The SDN automatically learns hierarchical features, salience maps, and mixture representations of different body parts. Pedestrian detection faces the challenges of background clutter and large variations of pedestrian appearance due to pose and viewpoint changes and other factors. One of our key contributions is to propose a Switchable Restricted Boltzmann Machine (SRBM) to explicitly model the complex mixture of visual variations at multiple levels. At the feature levels, it automatically estimates saliency maps for each test sample in order to separate background clutters from discriminative regions for pedestrian detection. At the part and body levels, it is able to infer the most appropriate template for the mixture models of each part and the whole body. We have devised a new generative algorithm to effectively pretrain the SDN and then fine-tune it with back-propagation. Our approach is evaluated on the Caltech and ETH datasets and achieves the state-of-the-art detection performance.

show abstract

Detection Evolution with Multi-order Contextual Co-occurrence

Cited by 78 publications

References 22 publications

Segmentation Driven Object Detection with Fisher Vectors

Segmentation Driven Object Detection with Fisher Vectors

Composite Discriminant Factor analysis

Switchable Deep Network for Pedestrian Detection

Contact Info

Product

Resources

About