<ul><li>Human image generation is a key focus in image synthesis with broad applications.</li><li>Direct Preference Optimization (DPO) is explored to improve the realism of generated images.</li><li>An enhanced DPO approach is proposed, incorporating high-quality real images as winning images.</li><li>The HG-DPO approach employs a curriculum learning framework for gradual improvements and personalized text-to-image tasks.</li></ul>

Boost Your Human Image Generation Model via Direct Preference Optimization

Discover more