Cooperative multi-scale Convolutional Neural Networks for person detection

Eisenbach, Markus; Seichter, Daniel; Wengefeld, Tim; Groß, Horst–Michael

doi:10.1109/ijcnn.2016.7727208

Cited by 16 publications

(6 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A solution for detecting pedestrians at different scales and evaluated on the Caltech data set by combining three CNNs was proposed in [9]. A cascade Aggregated Channel Features detector is used in [40] to create pedestrian candidate windows followed by a CNN-based classifier for assessment purposes on monocular Caltech and stereo ETH data sets.…”

Section: B Deep Learning Neural Network Modelsmentioning

confidence: 99%

Pedestrian Recognition Using Cross-Modality Learning in Convolutional Neural Networks

Pop

Rogozan

Nashashibi

et al. 2021

IEEE Intell. Transport. Syst. Mag.

View full text Add to dashboard Cite

The combination of multi-modal image fusion schemes with deep learning classification methods, and particularly with Convolutional Neural Networks (CNNs) has achieved remarkable performances in the pedestrian detection field. The late fusion scheme has significantly enhanced the performance of the pedestrian recognition task. In this paper, the late fusion scheme connected with CNN learning is deeply investigated for pedestrian recognition based on the Daimler stereo vision dataset. Thus, an independent CNN for each imaging modality (Intensity, Depth, and Optical Flow) is used before the fusion of the CNN's probabilistic output scores with a Multi-Layer Perceptron which provides the recognition decision. We propose four different learning patterns based on Cross-Modality deep learning of Convolutional Neural Networks: (1) a Particular Cross-Modality Learning; (2) a Separate Cross-Modality Learning; (3) a Correlated Cross-Modality Learning and ( 4) an Incremental Cross-Modality Learning model. Moreover, we also design a new CNN architecture, called LeNet+, which improves the classification performance not only for each modality classifier, but also for the multi-modality late-fusion scheme. Finally, we propose to learn the LeNet+ model with the incremental cross-modality approach using optimal learning settings, obtained with a K-fold Cross Validation pattern. This method outperforms the state-of-the-art classifier provided with Daimler datasets on both non-occluded and partially-occluded pedestrian tasks.

show abstract

Section: B Deep Learning Neural Network Modelsmentioning

confidence: 99%

Pedestrian Recognition Using Cross-Modality Learning in Convolutional Neural Networks

Pop

Rogozan

Nashashibi

et al. 2021

IEEE Intell. Transport. Syst. Mag.

View full text Add to dashboard Cite

show abstract

“…A combination of three CNNs to detect pedestrians at various scales was introduced on the same monocular vision data set [9]. A cascade Aggregated Channel Features detector is used in [10] to generate candidate pedestrian windows followed by a CNN-based classifier for verification purposes on monocular Caltech and stereo ETH data sets.…”

Section: Related Workmentioning

confidence: 99%

Detection of Pedestrian Actions Based on Deep Learning Approach

Pop¹

2019

Studia UBB Informatica

View full text Add to dashboard Cite

The pedestrian detection has attracted considerable attention from research due to its vast applicability in the field of autonomous vehicles. In the last decade, various investigations were made to find an optimal solution to detect the pedestrians, but less of them were focused on detecting and recognition the pedestrian's action. In this paper, we converge on both issues: pedestrian detection and pedestrian action recognize at the current detection time (T=0) based on the JAAD dataset, employing deep learning approaches. We propose a pedestrian detection component based on Faster R-CNN able to detect the pedestrian and also recognize if the pedestrian is crossing the street in the detecting time. The method is in contrast with the commonly pedestrian detection systems, which only discriminate between pedestrians and non-pedestrians among other road users.

show abstract

“…This experiment concentrated on the detection of small scale pedestrians on the Caltech data set. A combination of three CNNs to detect pedestrians at distinct scales was proposed on the same monocular vision data set [9]. A cascade Aggregated Channel Features detector is utilized in [10] to engender candidate pedestrian windows followed by a CNN-based classifier for checking purposes on monocular Caltech and stereo ETH data sets.…”

Section: Previous Workmentioning

confidence: 99%

Pedestrian recognition through different cross-modality deep learning methods

Pop

Rogozan

Nashashibi

et al. 2017

2017 IEEE International Conference on Vehicular Electronics and Safety (ICVES)

View full text Add to dashboard Cite

Abstract-A wide variety of approaches have been proposed for pedestrian detection in the last decay and it still remains an open challenge due to its outstanding importance in the field of automotive. In recent years, deep learning classification methods, in particular convolutional neural networks, combined with multi-modality images applied on different fusion schemes have achieved great performances in computer vision tasks. For the pedestrian recognition task, the late-fusion scheme outperforms the early and intermediate integration of modalities. In this paper, we focus on improving and optimizing the late-fusion scheme for pedestrian classification on the Daimler stereo vision data set. We propose different training methods based on Cross-Modality deep learning of Convolutional Neural Networks (CNNs): (1) a correlated model, (2) an incremental model and, (3) a particular cross-modality model, where each CNN is trained on one modality, but tested on a different one. The experiments show that the incremental cross-modality deep learning of CNNs achieves the best performances. It improves the classification performances not only for each modality classifier, but also for the multi-modality late-fusion scheme. The particular cross-modality model is a promising idea for automated annotation of modality images with a classifier trained on a different modality and/or for cross-dataset training.

show abstract

Cooperative multi-scale Convolutional Neural Networks for person detection

Cited by 16 publications

References 45 publications

Pedestrian Recognition Using Cross-Modality Learning in Convolutional Neural Networks

Pedestrian Recognition Using Cross-Modality Learning in Convolutional Neural Networks

Detection of Pedestrian Actions Based on Deep Learning Approach

Pedestrian recognition through different cross-modality deep learning methods

Contact Info

Product

Resources

About