Person Re-Identification via Attention Pyramid

Chen, Guangyi; Gu, Tianpei; Lu, Jiwen; Bao, Jin-An; Zhou, Jie

doi:10.1109/tip.2021.3107211

Cited by 90 publications

(28 citation statements)

References 74 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, compared with other methods, the results show that the proposed LTReID achieves superior or competitive re trieval accuracies. Compared with other datasets, the MSMTl 7 [95], the LTReID still achieves higher retrieval accuracies than the other methods. This clearly demonstrates that the LTRelD is able to achieve a satisfactory generalization on the large-scale dataset.…”

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 87%

LTReID: Factorizable Feature Generation With Independent Components for Long-Tailed Person Re-Identification

Wang

Zhao

et al. 2023

IEEE Trans. Multimedia

View full text Add to dashboard Cite

With the rapid increase of large-scale and real-world person datasets, it is crucial to address the problem of long-tailed data distributions, i.e., head classes have large number of images while tail classes occupy extremely few samples. We observe that the imbalanced data distribution is likely to distort the overall feature space and impair the generalization capability of trained models. Nevertheless, this long-tailed problem has been rarely investigated in previous person Re-Identification (ReID) works. In this paper, we propose a novel Long-Tailed Re-Identification (LTReID) framework to simultaneously alleviate class-imbalance and hard-imbalance problems. Specifically, each real feature is decomposed into multiple independent components with two decorrelation losses. Then these components are randomly aggre gated to generate more fake features for tail classes than head ones, resulting in the class-balance between head and tail classes. For the hard-balance between easy and hard samples, we utilize adversarial learning to generate more hard features than easy ones. The proposed framework can be trained in an end-to-end manner and avoids increasing the space and time complexity of inference models. Moreover, comprehensive experiments are conducted on the four ReID datasets so as to validate the effectiveness of the overall framework and the advantage of each module. Our results show that when trained with either balanced or imbalanced datasets, the LTReID achieves superior performance over the state-of-the-art methods.

show abstract

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 87%

LTReID: Factorizable Feature Generation With Independent Components for Long-Tailed Person Re-Identification

Wang

Zhao

et al. 2023

IEEE Trans. Multimedia

View full text Add to dashboard Cite

show abstract

“…Therefore, this can be considered analogous to transfer learning experiments in context of neural networks. Therefore, our 62.9% Market-CUHK rank-1 "transfer" accuracy is significant when compared to 50.1% DukeMTMC-Market transfer in [19], as shown in Table 6. We also achieve mAP which is comparable or even higher than the transfer learning mAP of APNet-C.…”

Section: Generalization and Potential Application To Open-world Scenariomentioning

confidence: 84%

“…However, generalization of the results is challenging for DL-assisted re-ID. That is, models trained on specific datasets tend to perform poorly on other data, as can be illustrated by models trained on Market1501 to 98% rank-1 accuracy reaching only 38% accuracy on DukeMTMC and 51% vice versa [19]. The effects of negative transfer learning can be drastic due to significant variations in data contents, e.g., different clothes people wear, different camera types, different filming environments.…”

Section: Introductionmentioning

confidence: 99%

Combining human parsing with analytical feature extraction and ranking schemes for high-generalization person reidentification

Gabdullin¹

2022

Preprint

View full text Add to dashboard Cite

Person reidentification (re-ID) has been receiving increasing attention in recent years due to its importance for both science and society. Machine learning and particularly Deep Learning (DL) has become the main re-id tool that allowed researches to achieve unprecedented accuracy levels on benchmark datasets. However, there is a known problem of poor generalization of DL models. That is, models trained to achieve high accuracy on one dataset perform poorly on other ones and require re-training. To address this issue, we present a model without trainable parameters which shows great potential for high generalization. It combines a fully analytical feature extraction and similarity ranking scheme with DL-based human parsing used to obtain the initial subregion classification. We show that such combination to a high extent eliminates the drawbacks of existing analytical methods. We use interpretable color and texture features which have humanreadable similarity measures associated with them. To verify the proposed method we conduct experiments on Market1501 and CUHK03 datasets achieving competitive rank-1 accuracy comparable with that of DL-models. Most importantly we show that our method achieves 63.9% and 93.5% rank-1 cross-domain accuracy when applied to transfer learning tasks. It is significantly higher than previously reported 30-50% transfer accuracy. We discuss the potential ways of adding new features to further improve the model. We also show the advantage of interpretable features for constructing human-generated queries from verbal description to conduct search without a query image.

show abstract

“…Another multi-scale attention pyramid method to mitigate the scale challenge was presented in [169]. First features were divided into multiple local parts and then learned using attention mechanism.…”

Section: Attention-based Approachesmentioning

confidence: 99%

“…The scale differences are handled by various re-id solutions, however, the attention based approaches outperformed the rest of re-id solutions. The top three solutions either used the multi-scale attention pyramid [165] or divided the image into multiple local parts and then learnt the attention [169], or [171] extracted the holistc and local feature maps using multi-scale omni-bearing attention network. The re-id solutions that support multi-scale re-id learn the person features at multiple scales through multi-sized convolutional layers or branches.…”

Section: Deep Learning Conjecturementioning

confidence: 99%

Person Re-identification: A Retrospective on Domain Specific Open Challenges and Future Trends

Zahra¹,

Perwaiz²,

Shahzad³

et al. 2022

Preprint

View full text Add to dashboard Cite

Person re-identification (Re-ID) is one of the primary components of an automated visual surveillance system. It aims to automatically identify/search persons in a multi-camera network having non-overlapping field-of-views. Owing to its potential in various applications and research significance, a plethora of deep learning based re-Id approaches have been proposed in the recent years. However, there exist several vision related challenges, e.g., occlusion, pose scale & viewpoint variance, background clutter, person misalignment and cross-domain generalization across camera modalities, which makes the problem of re-Id still far from being solved. Majority of the proposed approaches directly or indirectly aim to solve one or multiple of these existing challenges. In this context, a comprehensive review of current re-ID approaches in solving theses challenges is needed to analyze and focus on particular aspects for further advancements. At present, such a focused review does not exist and henceforth in this paper, we have presented a systematic challenge-specific literature survey of 230+ papers between the years of 2015-21. For the first time a survey of this type have been presented where the person re-Id approaches are reviewed in such solution-oriented perspective. Moreover, we have presented several diversified prominent developing trends in the respective research domain which will provide a visionary perspective regarding ongoing person re-Id research and eventually help to develop practical real world solutions. Scope/Objective of the ReviewIn this paper, we have targeted the most popular challenges in person re-id to perform systematic challenge-wise review of the published approaches. In this context, we have collected papers from top conferences and journals for the years from 2015 to 2021. The progress in papers addressing each challenge and its influence on published results is

show abstract

Person Re-Identification via Attention Pyramid

Cited by 90 publications

References 74 publications

LTReID: Factorizable Feature Generation With Independent Components for Long-Tailed Person Re-Identification

LTReID: Factorizable Feature Generation With Independent Components for Long-Tailed Person Re-Identification

Combining human parsing with analytical feature extraction and ranking schemes for high-generalization person reidentification

Person Re-identification: A Retrospective on Domain Specific Open Challenges and Future Trends

Contact Info

Product

Resources

About