Complementary Data Augmentation for Cloth-Changing Person Re-Identification

Jia, Xuemei; Zhong, Xian; Ye, Mang; Liu, Wenxuan; Huang, Wenxin

doi:10.1109/tip.2022.3183469

Cited by 36 publications

(9 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Jin et al's work 22 centered on single-image person Re-ID, catering to scenarios that demand effective and prompt human identity matching. Jia et al 23 explored the problem of person Re-ID in cloth-changing contexts, where individuals of the same identity may have different clothes. However, most existing methods have predominantly concentrated on body size or contour sketches, neglecting the full potential of human semantic information and pedestrian characteristics before and after clothing changes.…”

Section: Cloth-changing Person Re-identificationmentioning

confidence: 99%

See 1 more Smart Citation

FaceCC: facial feature combination network for cloth-changing person re-identification

Wu,

Ding,

Zhang

2024

J. Electron. Imag.

View full text Add to dashboard Cite

Person re-identification (Re-ID) is a significant research field that seeks to enable the identification of individuals among different surveillance cameras, making it a fundamental technology in the field of computer vision. Despite its significance, current models addressing this problem encounter difficulties when tasked with re-identifying individuals over an extended period, owing to changes in appearance such as clothing and hairstyles. We propose an innovative Re-ID network that considers facial features to enhance the model training and to improve the Re-ID performance by extracting pertinent facial information. Specifically, we filter the face images in the training set and optimize the training dataset, enabling our model to focus on essential facial features. In addition to the global stream in the training process, we introduce a distinct facial stream to train the facial features, which is subsequently merged into the overall flow at the loss function stage to enable knowledge transfer between the two branches. Experimental results on different datasets have shown that our method performs much better than the previous state-of-the-art methods in Rank-1 and mAP metrics. This proposed approach therefore presents a innovative and promising solution for the research of cloth-changing person Re-ID, and the associated code and dataset are made available on https://github.com/ Titor99/FaceCC

show abstract

Section: Cloth-changing Person Re-identificationmentioning

confidence: 99%

“…Jin et al.’s work 22 centered on single-image person Re-ID, catering to scenarios that demand effective and prompt human identity matching. Jia et al 23 . explored the problem of person Re-ID in cloth-changing contexts, where individuals of the same identity may have different clothes.…”

Section: Related Workmentioning

confidence: 99%

FaceCC: facial feature combination network for cloth-changing person re-identification

Wu,

Ding,

Zhang

2024

J. Electron. Imag.

View full text Add to dashboard Cite

show abstract

“…Recently, with the rapid development of deep learning (Xie et al 2023;Yang et al 2023;Yuan et al 2023), research on coreset selection for deep learning has emerged, including geometry-based methods (Sener and Savarese 2017;Agarwal et al 2020), uncertainty-based methods (Coleman et al 2019), submodularity-based methods (Kothawade et al 2022;Rangwani et al 2021), gradient matching-based methods (Mirzasoleiman, Bilmes, and Leskovec 2020;Killamsetty et al 2021a) and others (Toneva et al 2018;Swayamdipta et al 2020). They typically rely on a pre-trained model to obtain information, e.g., features,…”

Section: Introductionmentioning

confidence: 99%

Contributing Dimension Structure of Deep Feature for Coreset Selection

Wan,

Wang,

Wang

et al. 2024

AAAI

View full text Add to dashboard Cite

Coreset selection seeks to choose a subset of crucial training samples for efficient learning. It has gained traction in deep learning, particularly with the surge in training dataset sizes. Sample selection hinges on two main aspects: a sample's representation in enhancing performance and the role of sample diversity in averting overfitting. Existing methods typically measure both the representation and diversity of data based on similarity metrics, such as L2-norm. They have capably tackled representation via distribution matching guided by the similarities of features, gradients, or other information between data. However, the results of effectively diverse sample selection are mired in sub-optimality. This is because the similarity metrics usually simply aggregate dimension similarities without acknowledging disparities among the dimensions that significantly contribute to the final similarity. As a result, they fall short of adequately capturing diversity. To address this, we propose a feature-based diversity constraint, compelling the chosen subset to exhibit maximum diversity. Our key lies in the introduction of a novel Contributing Dimension Structure (CDS) metric. Different from similarity metrics that measure the overall similarity of high-dimensional features, our CDS metric considers not only the reduction of redundancy in feature dimensions, but also the difference between dimensions that contribute significantly to the final similarity. We reveal that existing methods tend to favor samples with similar CDS, leading to a reduced variety of CDS types within the coreset and subsequently hindering model performance. In response, we enhance the performance of five classical selection methods by integrating the CDS constraint. Our experiments on three datasets demonstrate the general effectiveness of the proposed method in boosting existing methods.

show abstract

“…The existing methods generally adopt the encoder-decoder framework (Venugopalan et al 2015), where the encoder generates visual representation by receiving a set of consecutive frames as input, and the decoder generates captions via recurrent neural networks (RNNs) or Transformer (Hori et al 2017). Some efforts (Hu et al 2022a,b;Jia et al 2022; Liao Xu et al 2022) are developed to explore richer visual features. Prior works (Zhang and Peng 2019;Pan et al 2020;Zhang et al 2020;Li et al 2022c) enhance the spatiotemporal representations between objects, while others (Jin et al 2020;Gao et al 2022) tend to improve the architecture of the decoder to obtain better linguistic representations.…”

Section: Introductionmentioning

confidence: 99%

Refined Semantic Enhancement towards Frequency Diffusion for Video Captioning

Xian

Chen

et al. 2023

AAAI

View full text Add to dashboard Cite

Video captioning aims to generate natural language sentences that describe the given video accurately. Existing methods obtain favorable generation by exploring richer visual representations in encode phase or improving the decoding ability. However, the long-tailed problem hinders these attempts at low-frequency tokens, which rarely occur but carry critical semantics, playing a vital role in the detailed generation. In this paper, we introduce a novel Refined Semantic enhancement method towards Frequency Diffusion (RSFD), a captioning model that constantly perceives the linguistic representation of the infrequent tokens. Concretely, a Frequency-Aware Diffusion (FAD) module is proposed to comprehend the semantics of low-frequency tokens to break through generation limitations. In this way, the caption is refined by promoting the absorption of tokens with insufficient occurrence. Based on FAD, we design a Divergent Semantic Supervisor (DSS) module to compensate for the information loss of high-frequency tokens brought by the diffusion process, where the semantics of low-frequency tokens is further emphasized to alleviate the long-tailed problem. Extensive experiments indicate that RSFD outperforms the state-of-the-art methods on two benchmark datasets, i.e., MSR-VTT and MSVD, demonstrate that the enhancement of low-frequency tokens semantics can obtain a competitive generation effect. Code is available at https://github.com/lzp870/RSFD.

show abstract

Complementary Data Augmentation for Cloth-Changing Person Re-Identification

Cited by 36 publications

References 72 publications

FaceCC: facial feature combination network for cloth-changing person re-identification

FaceCC: facial feature combination network for cloth-changing person re-identification

Contributing Dimension Structure of Deep Feature for Coreset Selection

Refined Semantic Enhancement towards Frequency Diffusion for Video Captioning

Contact Info

Product

Resources

About