Pedestrian detection aided by deep learning semantic tasks

@article{Tian2015PedestrianDA,
  title={Pedestrian detection aided by deep learning semantic tasks},
  author={Yonglong Tian and Ping Luo and Xiaogang Wang and Xiaoou Tang},
  journal={2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2015},
  pages={5079-5087}
}
Deep learning methods have achieved great successes in pedestrian detection, owing to its ability to learn discriminative features from raw pixels. However, they treat pedestrian detection as a single binary classification task, which may confuse positive with hard negative samples (Fig.1 (a)). To address this ambiguity, this work jointly optimize pedestrian detection with semantic tasks, including pedestrian attributes (e.g. `carrying backpack') and scene attributes (e.g. `vehicle', `tree… Expand
Deep Learning Strong Parts for Pedestrian Detection
TLDR
This work proposes DeepParts, which consists of extensive part detectors that can detect pedestrian by observing only a part of a proposal, and yields a new state-of-the-art miss rate of 11:89%, outperforming the second best method by 10%. Expand
Attribute-Aware Pedestrian Detection in a Crowd
TLDR
An attribute-aware pedestrian detector to explicitly model people's semantic attributes in a high-level feature detection fashion is proposed and outperforms state-of-the-art methods at a large margin on pedestrian detection. Expand
Scale-Adaptive Deconvolutional Regression Network for Pedestrian Detection
TLDR
This paper proposes a novel Scale-Adaptive Deconvolutional Regression (SADR) network, which can effectively detect pedestrians of various scales by flexibly choosing which feature layer to regress object locations according to the height of pedestrians, thus improving the detection accuracy significantly. Expand
Discriminative latent semantic feature learning for pedestrian detection
TLDR
A novel approach of learning discriminative features is proposed, addressing two main limitations of the methods in the literature by extending the learning process via a maximum margin criterion, in order to better encode discrim inative information directly in features. Expand
Multi-Grained Deep Feature Learning for Robust Pedestrian Detection
TLDR
This paper jointly train a multi-scale network and a human parsing network in a weakly supervised manner to predict pedestrians of particular scales with the most appropriate feature maps, by matching their receptive fields with the target sizes. Expand
Multi-Grained Deep Feature Learning for Pedestrian Detection
TLDR
This paper jointly train a scale-aware network and a human parsing network in a semi-supervised manner to predict pedestrians of particular scales using most appropriate feature maps, by matching their receptive field with the target sizes. Expand
FINE-TUNING DEEP LEARNING MODELS FOR PEDESTRIAN DETECTION
TLDR
Evaluating the performance of fine-tuned models with a variation of training data by comparing Faster Region-based Convolutional Neural Network (Faster R-CNN), Single Shot MultiBox Detector (SSD) Inception v2, and SSD Mobilenet v2 found that the precision and recall are more sensitive on the variation of the amount ofTraining data. Expand
MAPD: An improved multi-attribute pedestrian detection in a crowd
TLDR
An improved Multi-attribute pedestrian detection (MAPD) method, which coptimize person intra-class compactness and inter-class discrepancy, and proposes a novel Piecewise NMS algorithm to reduce false positive of small objects. Expand
Object Detection With Deep Learning: A Review
TLDR
This paper provides a review of deep learning-based object detection frameworks and focuses on typical generic object detection architectures along with some modifications and useful tricks to improve detection performance further. Expand
Pedestrian Attribute Detection using CNN
TLDR
This project implements traditional methods like multi-label SVM and softmax classifiers to predict attributes and compares their performance to ConvNets, and proposes two models ShallowMAR and ShallowSAR, where Shallow SAR treats the attributes independently, while Shallow MAR considers the attributes in a unified framework allowing to exploit the interdependence between attributes. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 43 REFERENCES
Multi-stage Contextual Deep Learning for Pedestrian Detection
TLDR
A new deep model is proposed that can jointly train multi-stage classifiers through several stages of back propagation and keeps the score map output by a classifier within a local region and uses it as contextual information to support the decision at the next stage. Expand
Robust Multi-resolution Pedestrian Detection in Traffic Scenes
TLDR
This paper takes pedestrian detection in different resolutions as different but related problems, and proposes a Multi-Task model to jointly consider their commonness and differences, which noticeably outperforms previous state-of-the-art techniques. Expand
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. Expand
Joint Deep Learning for Pedestrian Detection
TLDR
This paper forms these four important components in pedestrian detection into a joint deep learning framework and proposes a new deep network architecture that achieves a 9% reduction in the average miss rate compared with the current best-performing pedestrian detection approaches on the largest Caltech benchmark dataset. Expand
A discriminative deep model for pedestrian detection with occlusion handling
TLDR
A probabilistic pedestrian detection framework using a deformable part-based model to obtain the scores of part detectors and the visibilities of parts are modeled as hidden variables and a discriminative deep model is used for learning the visibility relationship among overlapping parts at multiple layers. Expand
Pedestrian Parsing via Deep Decompositional Network
TLDR
Experimental results show that the proposed new Deep Decompositional Network achieves better segmentation accuracy than the state-of-the-art methods on pedestrian images with or without occlusions. Expand
Pedestrian Detection with Unsupervised Multi-stage Feature Learning
TLDR
This work reports state-of-the-art and competitive results on all major pedestrian datasets with a convolutional network model that uses a few new twists, such as multi-stage features, connections that skip layers to integrate global shape information with local distinctive motif information. Expand
Facial Landmark Detection by Deep Multi-task Learning
TLDR
A novel tasks-constrained deep model is formulated, with task-wise early stopping to facilitate learning convergence and reduces model complexity drastically compared to the state-of-the-art method based on cascaded deep model. Expand
Local Decorrelation For Improved Pedestrian Detection
TLDR
Inspired by recent work on discriminative decorrelation of HOG features, an efficient feature transform that removes correlations in local neighborhoods is proposed, resulting in an overcomplete but locally decorrelated representation ideally suited for use with orthogonal decision trees. Expand
Pedestrian Attribute Recognition At Far Distance
TLDR
A new pedestrian attribute dataset is released, which is by far the largest and most diverse of its kind and it is shown that the large-scale dataset facilitates the learning of robust attribute detectors with good generalization performance. Expand
...
1
2
3
4
5
...