CharacterGen

CharacterGen is a framework designed for efficient 3D character generation from single images, utilizing a streamlined generation pipeline and an image-conditioned multi-view diffusion model. This model effectively calibrates input poses to a canonical form while retaining key attributes of the input image. The framework also includes a transformer-based, generalizable sparse-view reconstruction model, facilitating the creation of detailed 3D models from multi-view images. CharacterGen's approach has been thoroughly evaluated through quantitative and qualitative experiments, showing its proficiency in generating 3D characters with high-quality shapes and textures. CharacterGen is suitable for professionals in the animation and gaming industries who require high-quality 3D characters. It is particularly valuable for those who need to generate characters from single images, as it saves time and effort in the character creation process. By using CharacterGen, users can create detailed 3D models with high-quality textures, ready for downstream applications such as rigging and animation.