Multiple camera styles learning for unsupervised person re-identification

He, Linbing; Du, Haishun; Fu, Yiming; Ye, Yanfang

doi:10.1016/j.ijleo.2023.170718

Cited by 3 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among these approaches, [28], [29], [61], [62] are most similar to ours. Their methods aim to utilize camera labels to optimize the global model across cameras .…”

Section: Re-id With Auxiliary Informationsupporting

confidence: 55%

“…Comparison with UDA person Re-ID methods. We provide several recent unsupervised domain adaptation works for comparison, including MMT [46], SPCL [9], DARC [11], AWB [45], Fastreid [69], and CaCL [62]. Despite UDA-based methods utilizing external annotation to enhance Re-ID performance, they do not demonstrate significant improvement.…”

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

confidence: 99%

“…IICS [29] employs alternating intra-camera and inter-camera training to iteratively update the feature encoder. CaCL [62] establishes the camera-driven curriculum learning framework for progressively transferring knowledge from source to target domains.…”

Section: Re-id With Auxiliary Informationmentioning

confidence: 99%

“…w i is the cluster-wise weighting factor. We calculate w i follows [62]. The inter-camera contrastive loss enhances the person ID discrimination of the feature encoder within different cameras.…”

Section: Inter-camera Trainingmentioning

confidence: 99%

See 3 more Smart Citations

Camera-Aware Style Separation and Contrastive Learning for Unsupervised Person Re-Identification

Liang

Jin

et al. 2022

2022 IEEE International Conference on Multimedia and Expo (ICME)

View full text Add to dashboard Cite

Unsupervised person re-identification aims to retrieve images of a specified person without identity labels. Many recent unsupervised Re-ID approaches adopt clustering-based methods to measure cross-camera feature similarity to roughly divide images into clusters. They ignore the feature distribution discrepancy induced by camera domain gap, resulting in the unavoidable performance degradation. Camera information is usually available, and the feature distribution in the single camera usually focuses more on the appearance of the individual and has less intra-identity variance. Inspired by the observation, we introduce a Camera-Aware Label Refinement (CALR) framework that reduces camera discrepancy by clustering intracamera similarity. Specifically, we employ intra-camera training to obtain reliable local pseudo labels within each camera, and then refine global labels generated by inter-camera clustering and train the discriminative model using more reliable global pseudo labels in a self-paced manner. Meanwhile, we develop a cameraalignment module to align feature distributions under different cameras, which could help deal with the camera variance further. Extensive experiments validate the superiority of our proposed method over state-of-the-art approaches. The code is accessible at https://github.com/leeBooMla/CALR.

show abstract

“…Among these approaches, [28], [29], [61], [62] are most similar to ours. Their methods aim to utilize camera labels to optimize the global model across cameras .…”

Section: Re-id With Auxiliary Informationsupporting

confidence: 55%

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

confidence: 99%

Section: Re-id With Auxiliary Informationmentioning

confidence: 99%

Section: Inter-camera Trainingmentioning

confidence: 99%

See 2 more Smart Citations

Camera-Aware Style Separation and Contrastive Learning for Unsupervised Person Re-Identification

Liang

Jin

et al. 2022

2022 IEEE International Conference on Multimedia and Expo (ICME)

View full text Add to dashboard Cite

show abstract

“…Each approach can be applied according to the research domain being worked on. The supervised learning approach is known as the machine learning development process, which requires labeled data [5][6][7], while unsupervised learning is without labeled data [8][9][10]. Supervised learning is applied to develop classification models and requires labeled data.…”

Section: Introductionmentioning

confidence: 99%

Labeling Consistency Test of Multi-Label Data for Aspect and Sentiment Classification Using the Cohen Kappa Method

Chamid,

Widowati,

Kusumaningrum

2024

ISI

View full text Add to dashboard Cite

Classification is part of machine learning, and developing it requires labeled data. Most data is available in an unlabeled form. Data labeling is a step that researchers must take. Good labeled data will produce a good classification model. The data labeling process cannot be ignored and needs to be done carefully and consistently. Because the classification process requires well-labeled data that can be accounted for. In addition, good labeled data will produce a good classification model. The role of an expert (rater) is needed to label the data and ideally at least two experts. However, involving two raters will become a new problem because it is likely that the results of the inter-rater labeling will be different. We propose the Cohen Kappa method to overcome this problem. We used data from scraping user reviews of the Indonesian marketplace, there were 4.307. Based on the calculation results, Kappa=0.909 for aspect detection, Kappa=0.893 for sentiment classification, and Kappa=0.971 for class aspect. Based on the kappa value, the labeling results for aspect detection, sentiment classification and aspect class were declared "almost perfect agreement", so that the results of this research obtained labeled data that can be used for classification tasks, especially for developing aspect-based sentiment analysis models.

show abstract

Deep learning algorithms for person re-identification: sate-of-the-art and research challenges

Yadav

Vishwakarma

2023

Multimed Tools Appl

View full text Add to dashboard Cite

Multiple camera styles learning for unsupervised person re-identification

Cited by 3 publications

References 12 publications

Camera-Aware Style Separation and Contrastive Learning for Unsupervised Person Re-Identification

Camera-Aware Style Separation and Contrastive Learning for Unsupervised Person Re-Identification

Labeling Consistency Test of Multi-Label Data for Aspect and Sentiment Classification Using the Cohen Kappa Method

Deep learning algorithms for person re-identification: sate-of-the-art and research challenges

Contact Info

Product

Resources

About