Multi-scale cross-domain alignment for person image generation

Person image generation aims to generate images that maintain the original human appearance in different target poses.Recent works have revealed that the critical element in achieving this task is the alignment of appearance domain and pose domain.Previous alignment methods,such as appearance flow warping,correspondence learning and cross attention,often encounter challenges when it comes to producing fine texture details.These approaches suffer from limitations in accurately estimating appearance flows due to the lack of global receptive field.Alternatively,they can only perform cross-domain alignment on high-level feature maps with small spatial dimensions since the computational complexity increases quadratically with larger feature sizes.In this article,the significance of multi-scale alignment,in both low-level and high-level domains,for ensuring reliable cross-domain alignment of appearance and pose is demonstrated.To this end,a novel and effective method,named Multi-scale Crossdomain Alignment(MCA)is proposed.Firstly,MCA adopts global context aggregation transformer to model multi-scale interaction between pose and appearance inputs,which employs pair-wise window-based cross attention.Furthermore,leveraging the integrated global source information for each target position,MCA applies flexible flow prediction head and point correlation to effectively conduct warping and fusing for final transformed person image generation.Our proposed MCA achieves superior performance on two popular datasets than other methods,which verifies the effectiveness of our approach.

Research on the evolutionary game of the supervision of civil aviation dangerous goods transportation training

This study is carried out to promote the precise supervision of dangerous goods transportation training,improve the efficiency of civil aviation supervision and further ensure the safety of air transportation.First,from the perspective of behaviour interaction and interest demands,evolutionary game theory is used to construct a tripartite evolutionary game model of dangerous goods transportation training institutions,the Civil Aviation Administration of China(CAAC)and the public.Then,the evolutionary game equilibrium of the system is further analysed under the joint action of the three parties.Finally,the influences of important parameters of the model on the behavioural strategy selection of the three parties are investigated via MATLAB numerical simulation.The conclusions reveal that the system has three evolutionarily stable strategies under different parameters,namely(non-compliant training,supervision,non-participation in supervision),(non-compliant training,supervision,participation in supervision)and(compliant training,supervision,non-participation in supervision).Moreover,the CAAC supervision cost,the fine amount,the supervision cost of public participation,the supervision success rate and the reporting reward amount are the main parameters that affect the behavioural strategy selection of the tripartite game players.The conclusions and suggestions of this study provide a decision-making basis and guidance for the supervision and management of civil aviation dangerous goods transportation training.