Semantic Attribute Matching Networks
arXiv:1904.02969
Abstract
We present semantic attribute matching networks (SAM-Net) for jointly establishing correspondences and transferring attributes across semantically similar images, which intelligently weaves the advantages of the two tasks while overcoming their limitations. SAM-Net accomplishes this through an iterative process of establishing reliable correspondences by reducing the attribute discrepancy between the images and synthesizing attribute transferred images using the learned correspondences. To learn the networks using weak supervisions in the form of image pairs, we present a semantic attribute matching loss based on the matching similarity between an attribute transferred source feature and a warped target feature. With SAM-Net, the state-of-the-art performance is attained on several benchmarks for semantic matching and attribute transfer.
arXiv commentCVPR 2019
References in corpus (10)
- Texture Networks: Feed-forward Synthesis of Textures and Stylized Images
- Universal Correspondence Network
- Fast Patch-based Style Transfer of Arbitrary Style
- Neural Style Transfer: A Review
- Deep Photo Style Transfer
- Convolutional neural network architecture for geometric matching
- A Closed-form Solution to Photorealistic Image Stylization
- Learning Dense Correspondence via 3D-guided Cycle Consistency
- SCNet: Learning Semantic Correspondence
- FCSS: Fully Convolutional Self-Similarity for Dense Semantic Correspondence