Regionlets for Generic Object Detection

By Xiaoyu Wang, Ming Yang, Shenghuo Zhu, Yuanqing Lin. 2013

432 citations recorded by the source corpus.

Generic object detection is confronted by dealing with different degrees of variations in distinct object classes with tractable computations, which demands for descriptive and flexible object representations that are also efficient to evaluate for many locations. In view of this, we propose to model an object class by a cascaded boosting classifier which integrates various types of features from competing local regions, named as region lets. A region let is a base feature extraction region defined proportionally to a detection window at an arbitrary resolution (i.e. size and aspect ratio). These region lets are organized in small groups with stable relative positions to delineate fine grained spatial layouts inside objects. Their features are aggregated to a one-dimensional feature within one group so as to tolerate deformations. Then we evaluate the object bounding box proposal in selective search from segmentation cues, limiting the evaluation locations to thousands. Our approach significantly outperforms the state-of-the-art on popular multi-class detection benchmark datasets with a single method, without any contexts. It achieves the detection mean average precision of 41.7% on the PASCAL VOC 2007 dataset and 39.7% on the VOC 2010 for 20 object categories. It achieves 14.7% mean average precision on the Image Net dataset for 200 object categories, outperforming the latest deformable part-based model (DPM) by 4.7%.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Object Detection Networks on Convolutional Feature Maps (2015, methodology): “The classifier is often a linear SVM (possibly with a latent structure over the features) [2], a non-linear boosted classifier [3], or an additive kernel SVM [4].”
ImageNet Large Scale Visual Recognition Challenge (2014, neutral): “…Kuiyuan Yang‡, Chuntao Hong‡, Zheng Zhang‡ (Wang et al., 2014) NEC - - 19.6 NEC Labs America†, University of Missouri ‡ Xiaoyu Wang†, Miao Sun‡, Tianbao Yang†, Yuanqing Lin†, Tony X. Han‡, Shenghuo Zhu† (Wang et al., 2013) NUS 13.0 National University of Singapore Min Lin*, Qiang Chen*, Jian…”
Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition (2014, methodology): “The Regionlet method improves to 46.1% [8] by combining various features including conv5.”
Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation (2013, background): “Most of the competing submissions (OverFeat, NEC-MU, UvAEuvision, Toronto A, and UIUC-IFP) used convolutional neural networks, indicating that there is significant nuance in how CNNs can be applied to object detection, leading to greatly varying outcomes.”
Unsupervised Visual Representation Learning by Context Prediction (2015, neutral): “3.0 20.0 24.1 26.7 12.7 58.1 48.2 43.2 12.0 21.1 36.1 46.0 43.5 33.7 [8] w/o context 52.6 52.6 19.2 25.4 18.7 47.3 56.9 42.1 16.6 41.4 41.9 27.7 47.9 51.5 29.9 20.0 41.1 36.4 48.6 53.2 38.5 Regionlets[58] 54.2 52.0 20.3 24.0 20.1 55.5 68.7 42.6 19.2 44.2 49.1 26.6 57.0 54.5 43.4 16.4 36.6 37.7 59.4 ”
What Makes for Effective Detection Proposals? (2015, methodology): “The steady increase in complexity of the core classifiers has led to improved detection quality, but at the cost of significantly increased computation time per window [7], [8], [9], [10], [11].”
Deformable part models are convolutional neural networks (2014, neutral): “Regionlets [42] 65.”
Object-Proposal Evaluation Protocol is ‘Gameable’ (2015, background): “Types of Object Proposals: Object proposals can be broadly categorized into two categories: • Window scoring : In these methods, the space of all possible windows in an image is sampled to get a sub-set of the windows ( e.g ., via sliding window).”
A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection (2016, methodology): “While the R-CNN surpassed previous detectors [19,20] by a large margin, its speed is limited by the need for object proposal generation and repeated CNN evaluation.”
Part Detector Discovery in Deep Convolutional Neural Networks (2014, neutral): “Examples are the de-formable part model (DPM) [8], regionlets [15], and poselets [16].”
Computational Baby Learning (2014, methodology): “All the variants of our framework strongly outperform the methods [20] [26] [11] based on hand-crafted features learning and deformable part models.”
Domain Adaptation for Visual Applications: A Comprehensive Survey (2017, methodology): “rs trained on the features extracted from labeled bounding boxes. At test time, the classiﬁer was used to decide if a region of interest obtained by sliding windows or generic window selection models [192, 193, 194] contains the object or not. Therefore, considering the window selection mechanism as”
End-to-end integration of a Convolutional Network, Deformable Parts Model and non-maximum suppression (2014, neutral): “ 23.0 20.0 24.1 26.7 12.7 58.1 48.2 43.2 12.0 21.1 36.1 46.0 43.5 33.7 HSC-dpm [12] 32.2 58.3 11.5 16.3 30.6 49.9 54.8 23.5 21.5 27.7 34.0 13.7 58.1 51.6 39.9 12.4 23.5 34.4 47.4 45.2 34.3 Regionlets [17] 54.2 52.0 20.3 24.0 20.1 55.5 68.7 42.6 19.2 44.2 49.1 26.6 57.0 54.5 43.4 16.4 36.6 37.7 59.4 ”
Weakly Supervised Object Localization with Multi-Fold Multiple Instance Learning (2015, neutral): “On this dataset, the highest fully supervised detection result without using auxiliary data is 39.7% mAP [55].”
How good are detection proposals, really? (2014, methodology): “It is interesting to note that the current two top performing detection methods for Pascal VOC [11] use detection proposals [15, 32].”
Improving object detection with deep convolutional networks via Bayesian optimization and structured prediction (2015, methodology): “[18] proposed the R-CNN following the “recognition using regions” paradigm [19], which also inspired several previous state-of-the-art methods [44, 45].”
Learning Complexity-Aware Cascades for Deep Pedestrian Detection (2015, methodology): “This is usually addressed with object proposal mechanisms [12, 33, 15] that pre-select the most promising image patches.”
Real-Time, Cloud-Based Object Detection for Unmanned Aerial Vehicles (2017, methodology): “Most recent object detectors in the computer vision field use one of the objectness estimation techniques (or object proposal methods) for reducing computation instead of using brute-force sliding windows that run detectors at every possible image location [13], [26].”
Multi-instance object segmentation with occlusion handling (2015, methodology): “The “Regionlet” algorithm [34] detects an object by learning a cascaded boosting classifier with the most discriminative features extracted from subparts of regions, i.”