CVMay 25, 2017

Pose Guided Person Image Generation

arXiv:1705.09368v6889 citations
Originality Incremental advance
AI Analysis

This addresses the need for realistic person image synthesis in applications like virtual try-on or surveillance, though it builds incrementally on prior pose-guided generation methods.

The paper tackles the problem of generating person images in arbitrary poses from a single input image and a target pose, achieving high-quality results with convincing details on both re-identification and fashion photo datasets.

This paper proposes the novel Pose Guided Person Generation Network (PG$^2$) that allows to synthesize person images in arbitrary poses, based on an image of that person and a novel pose. Our generation framework PG$^2$ utilizes the pose information explicitly and consists of two key stages: pose integration and image refinement. In the first stage the condition image and the target pose are fed into a U-Net-like network to generate an initial but coarse image of the person with the target pose. The second stage then refines the initial and blurry result by training a U-Net-like generator in an adversarial way. Extensive experimental results on both 128$\times$64 re-identification images and 256$\times$256 fashion photos show that our model generates high-quality person images with convincing details.

Code Implementations2 repos
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes