VR-PROUD: Vehicle Re-identification using PROgressive Unsupervised Deep architecture

Bashir, Raja Muhammad Saad; Shahzad, Muhammad; Fraz, Muhammad Moazam

doi:10.1016/j.patcog.2019.01.008

Cited by 52 publications

(28 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…training requires a lot of labeled data. While unsupervised learning techniques can potentially cope with such issues by drawing inference directly from the unlabeled input data [75], and have been effectively employed in the context of person re-identification [76]- [78]. Deng et al [77] presented an unsupervised approach for image to image cross domain adaption using the self-similarity and domain-dissimilarity in the training, they used the similarity preserving GANs consisting of the Siamese Neural Networks using the contrastive loss for the re-identification purpose.…”

Section: ) Vehicle Re-identification Methods Based On Unsupervised Lmentioning

confidence: 99%

“…Some researchers have applied unsupervised methods to vehicle re-identification. A progressive two step cascaded framework was presented in [75], which essentially formulated the whole vehicle re-identification problem into an unsupervised learning paradigm, it combined a CNN architecture for feature extraction and an unsupervised technique to enable self-paced progressive learning, it also incorporated the contextual information into the proposed progressive framework that significantly improved the convergence of the learned algorithm. Marín-Reyes et al [79] applied the method in [80] to the vehicle re-identification task to create an annotation in an unsupervised manner, along with exploiting visual tracking to produce a weakly labeled training set.…”

Section: ) Vehicle Re-identification Methods Based On Unsupervised Lmentioning

confidence: 99%

“…Rank measures the similarity of a test to its owned class [75]. The value of rank-m is the probability of correct results in images with the highest confidence in the search results, a higher Rank indicates better performance of the model.…”

Section: ) Rankmentioning

confidence: 99%

“…Average precision (AP) measures how well the model judges the results on a single query image, while Mean average precision(mAP) measures how well the model judges the results on all query images. mAP is the average of all the AP, AP and mAP can be calculated as follows [75].…”

Section: ) Mapmentioning

confidence: 99%

See 3 more Smart Citations

A Survey of Vehicle Re-Identification Based on Deep Learning

2019

View full text Add to dashboard Cite

Vehicle re-identification is one of the core technologies of intelligent transportation systems, and it is crucial for the construction of smart cities. With the rapid development of deep learning, vehicle re-identification technologies have made significant progress in recent years. Therefore, making a comprehensive survey about the vehicle re-identification methods based on deep learning is quite indispensable. There are mainly five types of deep learning-based methods designed for vehicle re-identification, i.e. methods based on local features, methods based on representation learning, methods based on metric learning, methods based on unsupervised learning, and methods based on attention mechanism. The major contributions of our survey come from three aspects. First, we give a comprehensive review of the current five types of deep learning-based methods for vehicle re-identification, and we further compare them from characteristics, advantages, and disadvantages. Second, we sort out vehicle public datasets and compare them from multiple dimensions. Third, we further discuss the challenges and possible research directions of vehicle re-identification in the future based on our survey. INDEX TERMS Deep learning, intelligent transportation system, vehicle re-identification, vehicle public datasets.

show abstract

Section: ) Vehicle Re-identification Methods Based On Unsupervised Lmentioning

confidence: 99%

Section: ) Vehicle Re-identification Methods Based On Unsupervised Lmentioning

confidence: 99%

Section: ) Rankmentioning

confidence: 99%

Section: ) Mapmentioning

confidence: 99%

See 2 more Smart Citations

A Survey of Vehicle Re-Identification Based on Deep Learning

2019

View full text Add to dashboard Cite

show abstract

“…Currently, the popular solution always explores the unsupervised reID studies relying on the pseudolabels assigned by the clusters from one view [8], which neglects the potential labels from multi-views [9,10]. For instance, VR-PROUD [11] employs K-means algorithm to obtain cluster centroids which are regarded as labels for related samples. UDAR [12] iteratively guesses the unlabeled target data with DBSCAN and then the guessed labels are employed to train the model for unsupervised domain adaptive reID tasks.…”

Section: Introductionmentioning

confidence: 99%

Learning Multiple Semantic Knowledge For Cross-Domain Unsupervised Vehicle Re-Identification

Wang

Peng

Jiang

et al. 2021

2021 IEEE International Conference on Multimedia and Expo (ICME)

View full text Add to dashboard Cite

Unsupervised Vehicle re-identification (reID) aims at searching the similar vehicles' images from large unlabelled datasets captured in a multiple camera network, which is still a challenging task. In this paper, a multiple semantic knowledge learning approach is proposed to exploit the potential similarity of unlabeled samples, which builds multiple clusters from different views automatically with different cues. Specially, different from some existing works focus on the knowledge of one view, for each vehicle in the target domain, different semantic knowledge could be learned with the proposed focal drop network and several different labels can be assigned according these knowledge, which would be employed to train the vehicle reID model jointly. In addition, due to the unreliability of pseudo labels assigned by the clustering, the hard triplet center loss is proposed to take the difference of intra-cluster and inter-cluster into consideration for better training the unsupervised framework to adapt the unknown domain. Comprehensive experimental results clearly demonstrate that our method achieves excellent performance on both VehicleID dataset and VeRi-776 dataset.

show abstract

A new comparison framework to survey neural networks‐based vehicle detection and classification approaches

Hashemi

Emami

Sangar

2021

Int J Communication

View full text Add to dashboard Cite

Summary The vehicle detection and classification (VDC) problem has received much attention recently due to the increased security threats and the need to develop intelligent transportation systems. A large number of approaches have been proposed for the VDC problem using neural networks. To determine how neural networks‐based approaches have developed for the VDC in recent years, this paper surveys the VDC approaches through a literature review with the range Jan. 2012 through Apr. 2021. To do this, we introduce a new comparison framework to classify and compare the VDC approaches. Our proposed framework is composed of nine comparison dimensions: input data type, vehicle type, scale, scope, dynamicity, vehicle detection method, vehicle classification method, application, and evaluation method. Next, using the proposed framework, we discuss the evolution of the VDC approaches and identify several open issues that have emerged in the field. This paper provides a guide for researchers to use or design robust VDC systems with proper characteristics based on their needs.

show abstract

VR-PROUD: Vehicle Re-identification using PROgressive Unsupervised Deep architecture

Cited by 52 publications

References 16 publications

A Survey of Vehicle Re-Identification Based on Deep Learning

A Survey of Vehicle Re-Identification Based on Deep Learning

Learning Multiple Semantic Knowledge For Cross-Domain Unsupervised Vehicle Re-Identification

A new comparison framework to survey neural networks‐based vehicle detection and classification approaches

Contact Info

Product

Resources

About