2 papers
cs.CV2023
Shadows Don't Lie and Lines Can't Bend! Generative Models don't know Projective Geometry...for now
Ayush Sarkar, Hanlin Mai, Amitabh Mahapatra +3
Generative models can produce impressively realistic images. This paper demonstrates that generated images have geometric features different from those of real images. We build a s…
cs.CV2018
Out of the Box: Reasoning with Graph Convolution Nets for Factual Visual Question Answering
Medhini Narasimhan, Svetlana Lazebnik, Alexander G. Schwing
Accurately answering a question about a given image requires combining observations with general knowledge. While this is effortless for humans, reasoning with general knowledge re…