Local Feature-Aware Siamese Matching Model for Vehicle Re-Identification

Wang, Honglie; Sun, Shouqian; Zhou, Lunan; Guo, Lilin; Min, Xin; Li, Chao

doi:10.3390/app10072474

Cited by 7 publications

(4 citation statements)

References 39 publications

(55 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarity learning started as signature verification [18], and nowadays it is also largely used in the area of object re-identification (ReID). Face ReID is a wellresearched topic [45], [42], but also vehicle ReID has recently become a target for researchers [47], [110]. Tracking and object ReID face similar challenges and benefit from learning metric spaces that handle various appearance distortions of the object of interest [111].…”

Section: Discussionmentioning

confidence: 99%

Siamese Visual Object Tracking: A Survey

Ondrašovič

Tarábek

2021

IEEE Access

View full text Add to dashboard Cite

Object tracking belongs to active research areas in computer vision. We are interested in matching-based trackers exploiting deep machine learning known as Siamese trackers. Their powerful capabilities stem from similarity learning. This tracking paradigm is promising due to its inherent balance between performance and efficiency, so trackers of this type are suitable for real-time generic object tracking. There is an upsurge in research interest in Siamese trackers and the lack of available specialized surveys in this category. In this survey, we aim to identify and elaborate on the most significant challenges the Siamese trackers face. Our goal is to answer what design decisions the authors made and what problems they attempted to solve in the first place. We thus perform an in-depth analysis of the core principles on which Siamese trackers operate with a discussion of incentives behind them. Besides, we provide an up-todate qualitative and quantitative comparison of the prominent Siamese trackers on established benchmarks. Among other things, we discuss current trends in developing Siamese trackers. Our survey could help absorb the details about the underlying principles of Siamese trackers and the challenges they face.INDEX TERMS visual object tracking, deep learning, Siamese neural networks, similarity learning, fully convolutional networks

show abstract

Section: Discussionmentioning

confidence: 99%

Siamese Visual Object Tracking: A Survey

Ondrašovič

Tarábek

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The proposed RNN-based module models can effectively capture subtle visual appearance cues, such as paint and windshield stickers. In [3] , a local feature-aware model for vehicle re-identification was proposed to focus on learning discriminating parts that differ among vehicles. However, their model did not perform well under dim illumination conditions.…”

Section: Introductionmentioning

confidence: 99%

Impact of Color Space and Color Resolution on Vehicle Recognition Models

Ghanem,

Holliman

2024

J. Imaging

View full text Add to dashboard Cite

In this study, we analyze both linear and nonlinear color mappings by training on versions of a curated dataset collected in a controlled campus environment. We experiment with color space and color resolution to assess model performance in vehicle recognition tasks. Color encodings can be designed in principle to highlight certain vehicle characteristics or compensate for lighting differences when assessing potential matches to previously encountered objects. The dataset used in this work includes imagery gathered under diverse environmental conditions, including daytime and nighttime lighting. Experimental results inform expectations for possible improvements with automatic color space selection through feature learning. Moreover, we find there is only a gradual decrease in model performance with degraded color resolution, which suggests the need for simplified data collection and processing. By focusing on the most critical features, we could see improved model generalization and robustness, as the model becomes less prone to overfitting to noise or irrelevant details in the data. Such a reduction in resolution will lower computational complexity, leading to quicker training and inference times.

show abstract

“…The topic of vehicle re-identification has been studied extensively in the literature. In [ 1 ], the authors proposed a local feature-aware model for vehicle re-identification. Given multi-view images of a target vehicle, their model focuses on learning informative parts that are most likely to differ among vehicles.…”

Section: Introductionmentioning

confidence: 99%

Decision-Based Fusion for Vehicle Matching

Ghanem

Kerekes

Tokola

2022

Sensors

View full text Add to dashboard Cite

In this work, a framework is proposed for decision fusion utilizing features extracted from vehicle images and their detected wheels. Siamese networks are exploited to extract key signatures from pairs of vehicle images. Our approach then examines the extent of reliance between signatures generated from vehicle images to robustly integrate different similarity scores and provide a more informed decision for vehicle matching. To that end, a dataset was collected that contains hundreds of thousands of side-view vehicle images under different illumination conditions and elevation angles. Experiments show that our approach could achieve better matching accuracy by taking into account the decisions made by a whole-vehicle or wheels-only matching network.

show abstract

Local Feature-Aware Siamese Matching Model for Vehicle Re-Identification

Cited by 7 publications

References 39 publications

Siamese Visual Object Tracking: A Survey

Siamese Visual Object Tracking: A Survey

Impact of Color Space and Color Resolution on Vehicle Recognition Models

Decision-Based Fusion for Vehicle Matching

Contact Info

Product

Resources

About