Person Search

Person Search

person search旨在从一个原始的场景中找到特定的车辆,和person re-id是不同的,re-id依赖于crop的图片,search的任务更加具有现实意义。其实search的任务就是 detection + re-id


search可用的数据集

PRW dataset

Zheng, Liang, et al. “Person re-identification in the wild.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017.

CUHK-SYSU dataset:

Xiao, Tong, et al. “Joint detection and identification feature learning for person search.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017.

LSPS dataset:
Zhong, Yingji, Xiaoyu Wang, and Shiliang Zhang. “Robust Partial Matching for Person Search in the Wild.” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020.


Paper

主流的方法主要分为同步和异步:

end-to-end:

[1] Liu, Hao, et al. “Neural person search machines.” Proceedings of the IEEE International Conference on Computer Vision. 2017.

[2] Lan, Xu, Xiatian Zhu, and Shaogang Gong. “Person search by multi-scale matching.” Proceedings of the European Conference on Computer Vision (ECCV). 2018.

[3] Chang, Xiaojun, et al. “RCAA: Relational context-aware agents for person search.” Proceedings of the European Conference on Computer Vision (ECCV). 2018.

[4] He, Zhenwei, and Lei Zhang. “End-to-end detection and re-identification integrated net for person search.” Asian Conference on Computer Vision. Springer, Cham, 2018.

[5] Liu, Hong, et al. “A discriminatively learned feature embedding based on multi-loss fusion for person search.” 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2018.

[6] Yan, Yichao, et al. “Learning context graph for person search.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2019.

[7] Munjal, Bharti, et al. “Query-guided end-to-end person search.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2019.

[8] Zhai, Sulan, et al. “FMT: fusing multi-task convolutional neural network for person search.” Multimedia Tools and Applications 78.22 (2019): 31605-31616.

[9] Munjal, Bharti, Fabio Galasso, and Sikandar Amin. “Knowledge distillation for end-to-end person search.” arXiv preprint arXiv:1909.01058 (2019).

[10] Xiao, Jimin, et al. “IAN: the individual aggregation network for person search.” Pattern Recognition 87 (2019): 332-340.

[11] …

[12]Hong, Zheran, et al. “Scale Voting With Pyramidal Feature Fusion Network for Person Search.” IEEE Access 7 (2019): 139692-139702.

[13]Xue, Rui, et al. “Person Search with Joint Detection, Segmentation and Re-identification.” International Conference on Human Centered Computing. Springer, Cham, 2019.

[14]Loesch, Angelique, Jaonary Rabarisoa, and Romaric Audigier. “End-To-End Person Search Sequentially Trained On Aggregated Dataset.” 2019 IEEE International Conference on Image Processing (ICIP). IEEE, 2019.

[15]Lu, Yan, et al. “Dhff: Robust Multi-Scale Person Search by Dynamic Hierarchical Feature Fusion.” 2019 IEEE International Conference on Image Processing (ICIP). IEEE, 2019.

[16] Ştefan, Liviu-Daniel, et al. “Deep learning-based person search with visual attention embedding.” 2020 13th International Conference on Communications (COMM). IEEE, 2020.

[17] Chen, Zhicheng, et al. “FLAG: feature learning with additional guidance for person search.” The Visual Computer (2020): 1-9.

[18] Zheng, Dingyuan, et al. “Segmentation mask guided end-to-end person search.” Signal Processing: Image Communication (2020): 115876.

[19] Dai, Ju, et al. “Dynamic imposter based online instance matching for person search.” Pattern Recognition 100 (2020): 107120.

[20] Dong, Wenkai, et al. “Bi-Directional Interaction Network for Person Search.” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020.

[21] Di Chen14, Shanshan Zhang, et al. “Hierarchical Online Instance Matching for Person Search.” (2020).

[22] Chen, Di, et al. “Norm-Aware Embedding for Efficient Person Search.” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020.

[23] Zhang, Lei, et al. “Tasks Integrated Networks: Joint Detection and Retrieval for Image Search.” IEEE Transactions on Pattern Analysis and Machine Intelligence (2020).

[24]Shi, Wei, Hong Liu, and Mengyuan Liu. “Identity-sensitive loss guided and instance feature boosted deep embedding for person search.” Neurocomputing 415 (2020): 1-14.

[25]Tian, Kun, et al. “End-to-End Thorough Body Perception for Person Search.” AAAI. 2020.

[26]Chen, Haoran, et al. “Improved Model Structure with Cosine Margin OIM Loss for End-to-End Person Search.” International Conference on Multimedia Modeling. Springer, Cham, 2020.

[27]Liu, Qing, Keyang Cheng, and Bin Wu. “Person Search via Anchor-Free Detection and Part-Based Group Feature Similarity Estimation.” Chinese Conference on Pattern Recognition and Computer Vision (PRCV). Springer, Cham, 2020.

two-stage:

[1] Chen, Di, et al. “Person search via a mask-guided two-stream cnn model.” Proceedings of the European Conference on Computer Vision (ECCV). 2018.

[2]Han, Chuchu, et al. “Re-id driven localization refinement for person search.” Proceedings of the IEEE International Conference on Computer Vision. 2019.

[3] Wang, Cheng, et al. “TCTS: A Task-Consistent Two-Stage Framework for Person Search.” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020.

[4] Yao, Hantao, and Changsheng Xu. “Joint Person Objectness and Repulsion for Person Search.” arXiv preprint arXiv:2006.00155 (2020).

[5] Dong, Wenkai, et al. “Instance Guided Proposal Network for Person Search.” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020.

综述
Islam, Khawar. “Person search: New paradigm of person re-identification: A survey and outlook of recent works.” Image and Vision Computing 101 (2020): 103970.

Else:

Song, Binbin, et al. “Improving Person Search by Adaptive Feature Pyramid-based Multi-Scale Matching.” 2019 IEEE Visual Communications and Image Processing (VCIP). IEEE, 2019.

Huang, Zhongjie, Songlin Sun, and Yuhao Liu. “Person Search Based on Attention Mechanism.” 2019 19th International Symposium on Communications and Information Technologies (ISCIT). IEEE, 2019.

Li, Jianheng, et al. “Fast Person Search Pipeline.” 2019 IEEE International Conference on Multimedia and Expo (ICME). IEEE, 2019.

Gao, Cunyuan, et al. “Structure-aware person search with self-attention and online instance aggregation matching.” Neurocomputing 369 (2019): 29-38.

Zhang, Huimei, Changhong Chen, and Zongliang Gan. “Person Search Based on Improved Joint Learning Network.” Proceedings of the 3rd International Conference on Computer Science and Application Engineering. 2019.

Yao, Rui, et al. “GAN-based person search via deep complementary classifier with center-constrained Triplet loss.” Pattern Recognition (2020): 107350.

Chen, Di, et al. “Person Search by Separated Modeling and A Mask-Guided Two-Stream CNN Model.” IEEE Transactions on Image Processing 29 (2020): 4669-4682.

Yan, Lan, et al. “Weakly Supervised Person Search.” 2020 IEEE 7th International Conference on Data Science and Advanced Analytics (DSAA). IEEE, 2020.

Park, Jicheol, et al. “Learning Discriminative Part Features Through Attentions For Effective And Scalable Person Search.” 2020 IEEE International Conference on Image Processing (ICIP). IEEE, 2020.

Liu, Jiawei, et al. “Dual Context-Aware Refinement Network for Person Search.” Proceedings of the 28th ACM International Conference on Multimedia. 2020.

### 寻找与Person实体相关的LLM模型或资源 为了找到与Person实体相关的大型语言模型(LLM)或相关资源,可以从以下几个方面入手: #### 1. **基于开源社区的探索** 许多开源项目提供了针对特定领域优化的语言模型。例如,在Llama系列中,通过微调可以实现对Person实体更精确的理解和处理[^1]。具体来说,可以通过以下方法获取适合的模型: - 使用Hugging Face Model Hub搜索带有`person`标签或者专注于命名实体识别(NER)的任务模型。 - 探索专门用于人物分析的预训练模型,比如BioBERT、SciBERT等变体,这些可能已经经过调整以更好地理解个人身份。 #### 2. **定制化微调** 如果现有模型无法完全满足需求,则可考虑利用公开可用的数据集来进一步训练基础模型。例如,采用标注好的包含大量关于人的描述信息作为输入数据来进行fine-tuning操作。此过程中需要注意的是如何有效保存并应用已有的知识结构而不丢失泛化能力。 #### 3. **研究支持和服务平台** 对于那些希望深入了解背后机制以及寻求更多技术支持的研究人员而言,还可以联系专业的咨询服务团队获得更为详细的解答和支持[^2]。他们能够提供有关最佳实践案例分享和技术难题解决建议等方面的信息。 以下是Python脚本示例展示如何加载一个预先存在的transformer架构并通过自定义语料库对其进行再加工的过程: ```python from transformers import AutoTokenizer, AutoModelForSequenceClassification import torch tokenizer = AutoTokenizer.from_pretrained("bert-base-cased") model = AutoModelForSequenceClassification.from_pretrained("bert-base-cased") def fine_tune_model(training_data_path): # 加载本地文件中的训练样本... pass # 假设我们已经有了准备完毕后的dataset对象 training_dataset = ... optimizer = torch.optim.AdamW(model.parameters(), lr=5e-5) for epoch in range(epochs): for batch in training_loader: outputs = model(**batch) loss = outputs.loss loss.backward() optimizer.step() optimizer.zero_grad() ```
评论 5
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值