Make-A-Character 2: Animatable 3D Character Generation From a Single Image
This addresses the need for efficient 3D character creation in game development and digital human applications, building incrementally on prior work.
The paper tackles generating high-quality 3D characters from single portrait photos for games and digital humans, achieving the process in under 2 minutes and enabling real-time conversational animations.
This report introduces Make-A-Character 2, an advanced system for generating high-quality 3D characters from single portrait photographs, ideal for game development and digital human applications. Make-A-Character 2 builds upon its predecessor by incorporating several significant improvements for image-based head generation. We utilize the IC-Light method to correct non-ideal illumination in input photos and apply neural network-based color correction to harmonize skin tones between the photos and game engine renders. We also employ the Hierarchical Representation Network to capture high-frequency facial structures and conduct adaptive skeleton calibration for accurate and expressive facial animations. The entire image-to-3D-character generation process takes less than 2 minutes. Furthermore, we leverage transformer architecture to generate co-speech facial and gesture actions, enabling real-time conversation with the generated character. These technologies have been integrated into our conversational AI avatar products.