paper-with-me

Papers

Building Computationally Efficient and Well-Generalizing Person Re-Identification Models with Metric Learning

2020-03-17 · Submitted to International Conference on Pattern Recognition (ICPR 2020) 2020 3 · Vladislav Sovrasov, Dmitry Sidnev

This work considers the problem of domain shift in person re-identification.Being trained on one dataset, a re-identification model usually performs much worse on unseen data. Partially this gap is caused by the relatively small scale of person re-identification datasets (compared to face recognition ones, for instance), but it is also related to training objectives. We propose to use the metric learning objective, namely AM-Softmax loss, and some additional training practices to build well-generalizing, yet, computationally efficient models. We use recently proposed Omni-Scale Network (OSNet) architecture combined with several training tricks and architecture adjustments to obtain state-of-the art results in cross-domain generalization problem on a large-scale MSMT17 dataset in three setups: MSMT17-all->DukeMTMC, MSMT17-train->Market1501 and MSMT17-all->Market1501.

📄 PDF Abstract BibTeX arXiv:2003.07618

Code (1)

opencv/openvino_training_extensions 공식 구현 pytorch

Tasks

Domain GeneralizationFace RecognitionMetric LearningPerson Re-Identification

Similar Papers 제목 키워드 기반

Generalizing Person Re-Identification by Camera-Aware Invariance Learning and Cross-Domain Mixup

2020-08-01 · ECCV 2020 8 · Chuanchen Luo, Chunfeng Song, Zhao-Xiang Zhang

Despite the impressive performance under the single-domain setup, current fully-supervised models for person re-identification (re-ID) degrade significantly when deployed to an unseen domain. According to the characteris…

Person Re-Identification

Frustratingly Easy Person Re-Identification: Generalizing Person Re-ID in Practice

2019-05-09 · Jieru Jia, Qiuqi Ruan, Timothy M. Hospedales

Contemporary person re-identification (\reid) methods usually require access to data from the deployment camera network during training in order to perform well. This is because contemporary \reid{} models trained on one…

Person Re-Identification

Continual Person Identification using Footstep-Induced Floor Vibrations on Heterogeneous Floor Structures

2025-02-21 · Yiwen Dong, Hae Young Noh

Person identification is important for smart buildings to provide personalized services such as health monitoring, activity tracking, and personnel management. However, previous person identification relies on pre-collec…

Person Identification

Unveiling personnel movement in a larger indoor area with a non-overlapping multi-camera system

2021-04-10 · Ping Zhang, Zhenxiang Tao, Wenjie Yang, Minze Chen 외

Surveillance cameras are widely applied for indoor occupancy measurement and human movement perception, which benefit for building energy management and social security. To address the challenges of limited view angle of…

energy managementManagementPerson Re-Identification

Combined Depth Space based Architecture Search For Person Re-identification

2021-04-09 · CVPR 2021 1 · Hanjun Li, Gaojie Wu, Wei-Shi Zheng

Most works on person re-identification (ReID) take advantage of large backbone networks such as ResNet, which are designed for image classification instead of ReID, for feature extraction. However, these backbones may no…

image-classificationImage ClassificationPerson Re-IdentificationTriplet