InsHuman: Towards Natural and Identity-Preserving Human Insertion

TL;DR

InsHuman introduces a novel method for natural human insertion into images that preserves identity and adapts to backgrounds, addressing limitations of previous models and datasets.

Contribution

The paper proposes HBAF, FFIP, and BDP strategies to improve human insertion quality, identity preservation, and dataset realism in image editing.

Findings

InsHuman outperforms existing models in generating realistic, identity-preserving human insertions.

The proposed methods ensure coherent pose, count, and appearance alignment with backgrounds.

Experiments validate the effectiveness of InsHuman in producing high-quality images with preserved identities.

Abstract

Human insertion aims to naturally place specific individuals into a target background. Although existing image editing models may have such ability, they often produce failure cases, including inappropriate human pose in new background, inconsistent number of people, and modified facial identity. Moreover, publicly available human datasets often lack full-body portraits and realistic physical interaction between humans and their background. To address these challenges, we propose InsHuman for natural and identity-preserving human insertion. Specifically, we propose Human-Background Adaptive Fusion (HBAF), which detects foreground humans to obtain a binary mask and applies region-aware weighting to align the human regions between predicted and ground-truth latents, ensuring the person's pose, count, and overall appearance are coherently adapted to the target background.We further propose Face-to-Face ID-Preserving (FFIP), which detects and matches faces between the generated image and the source image in terms of face recognition features to enforce identity consistency for each face.In addition, we propose Bidirectional Data Pairing (BDP) strategy to construct BDP-InsHuman, a high-quality dataset with realistic human-background interactions. Experiments demonstrate that InsHuman achieves significant improvements in generating plausible images while keeping human identity unchanged.