SharinGAN: Combining Synthetic and Real Data for Unsupervised Geometry Estimation
This addresses the challenge of limited labeled real data for geometry estimation tasks in computer vision, though it appears incremental as it builds on existing unsupervised approaches.
The paper tackles the problem of unsupervised geometry estimation from single images by combining synthetic and real data, achieving significant improvements over state-of-the-art methods in surface normal estimation for human faces and monocular depth estimation for outdoor scenes.
We propose a novel method for combining synthetic and real images when training networks to determine geometric information from a single image. We suggest a method for mapping both image types into a single, shared domain. This is connected to a primary network for end-to-end training. Ideally, this results in images from two domains that present shared information to the primary network. Our experiments demonstrate significant improvements over the state-of-the-art in two important domains, surface normal estimation of human faces and monocular depth estimation for outdoor scenes, both in an unsupervised setting.