Robust Multi-resolution Pedestrian Detection in Traffic Scenes

By Junjie Yan, Xucong Zhang, Zhen Lei, Shengcai Liao, S. Li. 2013

185 citations recorded by the source corpus.

Subfields: Automotive 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 Automotive Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Real-Time Pedestrian Detection with Deep Network Cascades (2015, methodology): “Some methods focus on improving the base features used [6, 9, 10, 42], whereas others focus on the learning algorithms [13, 25], or other techniques such as incorporating Deformable Parts Models [33, 40] or using context [13, 29, 40, 41].”
Filtered channel features for pedestrian detection (2015, methodology): “We include for comparison the baselines mentioned in section 2.2, Roerei [1] the best known method trained without any Caltech images, MT-DPM [42] the best known method based on DPM, and SDN [21] the best known method using convolutional neural networks.”
Vehicle detection in aerial imagery : A small target detection benchmark (2016, result): “These bad results are consistent with previous works on pedestrian detection, in low resolution images which report a miss rate of more than 60% [85, 86] at 0.1 FFPI for images of 640x480 pixels.”
Pedestrian detection with a Large-Field-Of-View deep network (2015, methodology): “We believe, however, that some additional sources of context, such as the presence of a vehicle as in Yan et al [29] can further improve our proposed method.”
The Fastest Deformable Part Model for Object Detection (2014, methodology): “The proposed method is on par with the best performance method MT-DPM [33] and outperforms the ViolaJones style detector ACF [4] by 2%.”
Scale-Aware Fast R-CNN for Pedestrian Detection (2015, background): “Yan et al. [43] proposed to map the pedestrians in different resolutions to a common subspace to reduce the differences of local features.”
Single-Pedestrian Detection Aided by Multi-pedestrian Detection (2013, methodology): “approaches such as MultiResC [62], Rorei [5], MOCO [9], Crosstalk [19], MT-DPM [92], ACF [18], and ConvNet [70].”
Ten Years of Pedestrian Detection, What Have We Learned? (2014, methodology): “For pedestrian detection the results are competitive, but not salient (LatSvm [50,12], MultiResC [33], MT-DPM [39]).”
The CASIA NIR-VIS 2.0 Face Database (2013, background): “For example, inspired by the coupled spectral regression (CSR) in [5], [17] trains a scale robust pedestrian detector by joint learning from the samples with different scales.”
Using Machine Learning Safely in Automotive Software: An Assessment and Adaption of Software Process Requirements in ISO 26262 (2018, methodology): “The Deformable Part Model approach [62] has been successfully used this way for pedestrian detection [62].”
SAN: Learning Relationship between Convolutional Features for Multi-Scale Object Detection (2018, methodology): “A typical way of reducing the variance is partitioning samples by pose [18,20,34], rotation [36] or resolution [25,35].”
Occlusion-aware R-CNN: Detecting Pedestrians in a Crowd (2018, background): “A common paradigm [13,58,3,59,64] to address this problem is to train a pedestrian detector that exhaustively operates on the sub-images across all locations and scales.”
Local Decorrelation For Improved Pedestrian Detection (2014, background): “While in the past year the use of deep learning [25], motion features [27], and multi-resolution models [36] has brought down log-average miss rate (MR) to under 40% on the Caltech Pedestrian Dataset [10], LDCF reduces MR to under 25%.”
Object detection by labeling superpixels (2015, methodology): “The current publicly available state-of-the-art face detection [37, 5] and pedestrian detection [55, 57] methods are all based on sliding window.”