Source-free domain adaptive person re-identification(ReID)aims to address the cross-domain person ReID task with a well-trained source model,which solves the limitations of data privacy and transmission costs in real-...Source-free domain adaptive person re-identification(ReID)aims to address the cross-domain person ReID task with a well-trained source model,which solves the limitations of data privacy and transmission costs in real-world scenarios.Existing methods mainly generate pseudo labels for target data,which are unreliable because of distribution shifts.First,the ubiquitous difficult samples may lead to the ambiguity of the model prediction.Second,the source model may have a bias towards certain classes.To alleviate these two problems,we propose a source-free domain adaptive person ReID method via the uncertain label refinery(RULER).RULER consists of uncertainty-aware pseudo-labeling refinery(UPLR)and frequency-weighted contrastive learning(FCL).To reduce the ambiguity of predictions caused by sample label uncertainty,UPLR generates pseudo labels by clustering samples after multiple random dropouts and then integrates the results to obtain high-confidence pseudo labels.Moreover,FCL defines the frequency of each class as the sample weight and introduces a frequency-weighted contrastive loss to alleviate the class biases of the model.RULER improves the quality of pseudo labels and mitigates the source model′s bias towards certain classes.We achieve competitive results compared to state-of-the-art methods on both real-to-real and synthetic-to-real source-free domain adaptation scenarios,validating the effectiveness of RULER.展开更多
Presents a study which examined the structure-preserving algorithms to phase space volume for linear dynamical systems. Preservation of phase space volume for source-free dynamical systems; Description of a volume-pre...Presents a study which examined the structure-preserving algorithms to phase space volume for linear dynamical systems. Preservation of phase space volume for source-free dynamical systems; Description of a volume-preserving scheme for linear system with canonical form; Information on structure-preserving schemes for linear dynamical systems.展开更多
基金supported in part by the National Natural Science Foundation of China(Nos.62372003 and 62376004)the Natural Science Foundation of Anhui Province,China(Nos.2308085Y40 and 2208085J18)the University Synergy Innovation Program of Anhui Province,China(No.GXXT-2022-036).
文摘Source-free domain adaptive person re-identification(ReID)aims to address the cross-domain person ReID task with a well-trained source model,which solves the limitations of data privacy and transmission costs in real-world scenarios.Existing methods mainly generate pseudo labels for target data,which are unreliable because of distribution shifts.First,the ubiquitous difficult samples may lead to the ambiguity of the model prediction.Second,the source model may have a bias towards certain classes.To alleviate these two problems,we propose a source-free domain adaptive person ReID method via the uncertain label refinery(RULER).RULER consists of uncertainty-aware pseudo-labeling refinery(UPLR)and frequency-weighted contrastive learning(FCL).To reduce the ambiguity of predictions caused by sample label uncertainty,UPLR generates pseudo labels by clustering samples after multiple random dropouts and then integrates the results to obtain high-confidence pseudo labels.Moreover,FCL defines the frequency of each class as the sample weight and introduces a frequency-weighted contrastive loss to alleviate the class biases of the model.RULER improves the quality of pseudo labels and mitigates the source model′s bias towards certain classes.We achieve competitive results compared to state-of-the-art methods on both real-to-real and synthetic-to-real source-free domain adaptation scenarios,validating the effectiveness of RULER.
文摘Presents a study which examined the structure-preserving algorithms to phase space volume for linear dynamical systems. Preservation of phase space volume for source-free dynamical systems; Description of a volume-preserving scheme for linear system with canonical form; Information on structure-preserving schemes for linear dynamical systems.
