Problem
Research questions and friction points this paper is trying to address.
object placement
image composition
efficiency and effectiveness
Hybrid Generative-Discriminative Object Placement
๐ค AI Summary
ๆฌๆๆๅบไธ็งๅ็ๆๆนๆณ๏ผ้่ฟๅๅๅๅธ้็นๅนถ่ๅๅๆฏ่ๆฏ็นๅพ้ขๆตๅ็ๅๆฐๅๅฏ่ฝไฝ็ฝฎ๏ผไปฅๅนณ่กก็ฉไฝๆพ็ฝฎ็ๆ็ไธๆๆใ
๐ Abstract
As an important operation of image composition, object placement aims to predict the plausible placement (location, scale) for the inserted foreground object. Previous object placement methods can be divided into generative methods and discriminative methods, both of which cannot balance efficiency and effectiveness well. In this work, we propose a semi-generative method in the middle ground between them. In particular, we assign uniformly distributed anchors on the background. Then, we fuse foreground and background features to predict the rationality score for each anchor and predict plausible placement sets for positive anchors. Extensive experiments on the OPA dataset show that our method can strike a good balance between efficiency and effectiveness.
Innovation
Methods, ideas, or system contributions that make the work stand out.
semi-generative method
uniformly distributed anchors
foreground and background feature fusion
rationality score
๐ Similar Papers
No similar papers found.
