• Corpus ID: 195345825

Convolutional Channel Features For Pedestrian, Face and Edge Detection

@article{Yang2015ConvolutionalCF,
  title={Convolutional Channel Features For Pedestrian, Face and Edge Detection},
  author={Binh Yang and Junjie Yan and Zhen Lei and S. Li},
  journal={ArXiv},
  year={2015},
  volume={abs/1504.07339}
}
In this paper, we revisit the multiple channel features approach proposed by Dollár et al. [10,11], which has shown excellent performances in various computer vision tasks. Enlightened by the ConvNets, we introduce an extended version of multiple channel features called Convolutional Channel Features (CCF), which transfers low-level features from off-the-shelf ConvNet models to feed the boosting classifiers based on decision trees. With the combination of CNN features and decision trees, CCF… 

Figures and Tables from this paper

Pedestrian detection with motion features via two-stream ConvNets
TLDR
This work introduces transfer learning from multiple sources in the two-stream networks, which can transfer still image and motion features from ImageNet and an action recognition dataset respectively, to overcome the insufficiency of training data for convolutional neural networks in pedestrian datasets.
Multi-layer Feature Fusion and Selection from Convolutional Neural Networks for Texture Classification
TLDR
This paper aggregates CNN activations from different convolutional layers and encoding them into a single feature vector after applying a pooling operation, and involves a feature selection step that outperforms the state-of-the-art methods with a significant margin.
Joint Training of Cascaded CNN for Face Detection
TLDR
It is shown that the back propagation algorithm used in training CNN can be naturally used inTraining CNN cascade, and how jointly training can be conducted on naive CNN cascade and more sophisticated region proposal network (RPN) and fast R-CNN.
On the Discriminative Power of Learned vs. Hand-Crafted Features for Crowd Density Analysis
TLDR
A novel approach for crowd density classification is proposed, in which learned features substitute the commonly used handcrafted features, and the results demonstrate the effectiveness of learned features for crowddensity classification.
Minimal filtered channel features for pedestrian detection
This paper addresses the problem of efficient pedestrian detection using features that are extracted by convolving feature channels with a very small number of filters. The method uses as feature
Supervised Transformer Network for Efficient Face Detection
TLDR
A new cascaded Convolutional Neural Network, dubbed the name Supervised Transformer Network, to address the challenge of large pose variations in real-word face detection, which achieves state-of-the-art detection accuracies on several public benchmarks.
Scale-Aware Face Detection
TLDR
Scale-aware Face Detection (SAFD) is proposed to handle scale explicitly using CNN, and achieve better performance with less computation cost.
Pedestrian Detection for Autonomous Vehicle Using Multi-Spectral Cameras
TLDR
A novel instrument for pedestrian detection by combining stereo vision cameras with a thermal camera is presented, and it significantly outperforms the traditional histogram of oriented gradients features.
Looking at Pedestrians at Different Scales: A Multiresolution Approach and Evaluations
TLDR
This comprehensive evaluation of a multiresolution detector framework by training models at different sizes and demonstrating its effectiveness on a state-of-the-art pedestrian detector demonstrates meaningful improvement in detector performance.
...
...

References

SHOWING 1-10 OF 49 REFERENCES
Aggregate channel features for multi-view face detection
TLDR
Following the learning pipelines in Viola-Jones framework, the multi-view face detector using aggregate channel features shows competitive performance against state-of-the-art algorithms on AFW and FDDB test-sets, while runs at 42 FPS on VGA images.
Boosting Convolutional Features for Robust Object Proposals
TLDR
A boosting approach is proposed which directly takes advantage of hierarchical CNN features for detecting regions of interest fast and is demonstrated on ImageNet 2013 detection benchmark and compared with state-of-the-art methods.
Switchable Deep Network for Pedestrian Detection
TLDR
A new generative algorithm is devised to effectively pretrain the SDN and then fine-tune it with back-propagation to achieve the state-of-the-art detection performance.
Very Deep Convolutional Networks for Large-Scale Image Recognition
TLDR
This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
TLDR
This paper proposes a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%.
Generic Object Detection with Dense Neural Patterns and Regionlets
TLDR
Dense Neural Patterns, short for DNPs, are introduced, which are dense local features derived from discriminatively trained deep convolutional neural networks that can be easily plugged into conventional detection frameworks in the same way as other denseLocal features like HOG or LBP.
Learning and Transferring Mid-level Image Representations Using Convolutional Neural Networks
TLDR
This work designs a method to reuse layers trained on the ImageNet dataset to compute mid-level image representation for images in the PASCAL VOC dataset, and shows that despite differences in image statistics and tasks in the two datasets, the transferred representation leads to significantly improved results for object and action classification.
Taking a deeper look at pedestrians
TLDR
This paper analyses small and big convnets, their architectural choices, parameters, and the influence of different training data, including pretraining on surrogate tasks, and presents the best convnet detectors on the Caltech and KITTI dataset.
Integral Channel Features
TLDR
It is demonstrated that when designed properly, integral channel features not only outperform other features including histogram of oriented gradient (HOG), they also result in fast detectors when coupled with cascade classifiers.
Untangling Local and Global Deformations in Deep Convolutional Networks for Image Classification and Sliding Window Detection
TLDR
This work introduces epitomic con-volution as a building block alternative to the common convolution-MP cascade of DCNNs and develops an efficient DCNN sliding window object detector that employs explicit search over position, scale, and aspect ratio.
...
...