Robust object detection via soft cascade

By Lubomir D. Bourdev, Jonathan Brandt. 2005

388 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

What Makes for Effective Detection Proposals? (2015, background): “Each of the proposals has its own location bias to diversify the location of the proposals.”
A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection (2016, methodology): “Two main streams of research have been pursued to improve its speed: fast feature extraction [1,2] and cascade learning [14,17,15].”
DeepID-Net: multi-stage and deformable deep convolutional neural networks for object detection (2014, neutral): “i-stagetraining Motivation. Multi-stage classiﬁers have been widely used in object detection and achieved great success. With a cascaded structure, each classiﬁer processes a different subset of data [54, 15, 5, 19, 53]. However, these classiﬁers are usually trained sequentially without joint optimi”
Learning Complexity-Aware Cascades for Deep Pedestrian Detection (2015, methodology): “Early approaches used heuristics to find a cascade configuration of good trade-off between classification accuracy and complexity [32, 3, 35, 34].”
Discriminative local binary patterns for human detection in personal album (2008, background): “For comparison between several descriptors, it is a serious issue to optimally select the learning parameters such as the minimum detection rate in each cascade stage, which usually depend on feature type and need be empirically determined (see [3]).”
Revisiting RCNN: On Awakening the Classification Power of Faster RCNN (2018, background): “Soft cascades [1] improved [27] built each weak classifier based on the output of all previous classifiers.”
Cascade Ranking for Operational E-commerce Search (2017, methodology): “Cascade learning was rst applied with traditional classication and detection problems such as fast visual object detection in [4, 18, 20].”
Fast Multiclass Vehicle Detection on Aerial Images (2015, methodology): “To leverage this property for speed improvement, we form a soft cascade [12] from the weak classifiers.”
Scale-Aware Face Detection (2017, background): “Many works have been proposed to improve the Viola-Jones framework and achieves further improvements, such as local features [41, 20, 36], boosting algorithms [40, 21, 11], cascade structure [2] and multi-pose [22, 17, 12].”
Deep Learning for Generic Object Detection: A Survey (2018, background): “ssiﬁers at early stages discard large number of easy negative samples so that classiﬁers at latter stage can focus on handling more difﬁcult examples. Cascade has been widely used in object detection [71,20,155]. Two-stage object detection can be considered as cascade of two object detectors, the ﬁr”
Fast Feature Pyramids for Object Detection (2014, background): “Cascades [27], [47], [48], [49], [50], coarse-to-fine search [51], distance transforms [52], etc., all focus on optimizing classification speed given precomputed image features.”
Efficient Classification for Additive Kernel SVMs (2013, methodology): “This idea has been applied to face [24], [4] and pedestrian detection [64] to achieve an order of magnitude speedup in the overall detection time.”
Local Decorrelation For Improved Pedestrian Detection (2014, background): “Details, however, have evolved considerably; in particular, significant progress has been made on the feature representation [6] [9] [2] and cascade architecture [3] [8].”
Multi-view Face Detection Using Deep Convolutional Neural Networks (2015, methodology): “41] proposed to train a detector cascade for each view of the face and combined their results at the test time. Recently, [22] combined this method with integral channel features [3] and soft-cascade [1], and showed that by using 22 cascades, it is possible to obtain state-of-the-art performance for”
Poselets: Body part detectors trained using 3D human pose annotations (2009, methodology): “Poselets (this paper) Face detector of [2] Pedestrian detector of [3] Parts−based detector of [5]”
Aggregate channel features for multi-view face detection (2014, methodology): “Second difference is that soft-cascade [1] structure is used.”
Chained Cascade Network for Object Detection (2017, methodology): “Classifier chaining was effective in face detection [2, 35] and generic object detection [7] for hand crafted features.”