Pose-Normalized Image Generation for Person Re-identification

TL;DR

This paper introduces PN-GAN, a pose-normalization GAN that synthesizes realistic person images to improve re-identification by addressing pose variations and data scarcity, demonstrating strong, transferable features.

Contribution

The paper presents a novel GAN-based model for pose normalization in person re-identification, enabling scalable and transfer-friendly feature learning without additional dataset collection.

Findings

Synthesized images improve re-id feature robustness.

Features learned are strong and complementary to original image features.

Model generalizes well across different re-id datasets.

Abstract

Person Re-identification (re-id) faces two major challenges: the lack of cross-view paired training data and learning discriminative identity-sensitive and view-invariant features in the presence of large pose variations. In this work, we address both problems by proposing a novel deep person image generation model for synthesizing realistic person images conditional on the pose. The model is based on a generative adversarial network (GAN) designed specifically for pose normalization in re-id, thus termed pose-normalization GAN (PN-GAN). With the synthesized images, we can learn a new type of deep re-id feature free of the influence of pose variations. We show that this feature is strong on its own and complementary to features learned with the original images. Importantly, under the transfer learning setting, we show that our model generalizes well to any new re-id dataset without the need for collecting any training data for model fine-tuning. The model thus has the potential to make re-id model truly scalable.