Abstract
Attribute based person image generation is a problem which considers generating realistic person images using attributes like pose, gender, clothes, whether a bag is present or not etc. and it has wide variety of applications on computer vision. Realization of that generation process is quite difficult due to several reasons such as foreground/background, partial occlusion, stance of a person, camera angle and distance, complex relationships between attributes, unbalanced and poor quality data etc. Synthetic images are generated in related works which have worked on relatively easier datasets using less attributes with more complex models for more specific purposes. In this work, a more controversial goal was set up and a model named DCGAN-C is developed based on conditional generative adversarial networks and it can produce sythetic person images with both multi-class and multi-labels. Consequently, both quantitative and qualitative experiments were performed on the PA-100K dataset and the performance of the model was demonstrated.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
