GlassesGB: Controllable 2D GAN-Based Eyewear Personalization for 3D Gaussian Blendshapes Head Avatars
This work addresses the need for fine-grained, customizable eyewear in virtual reality try-on systems, representing an incremental advancement by combining existing techniques.
The paper tackles the problem of limited customization in virtual try-on systems for eyewear by proposing GlassesGB, a framework that integrates 2D generative adversarial networks with 3D Gaussian blendshapes to enable user-driven eyewear personalization for 3D head avatars in VR applications, achieving effective bridging between 2D design and 3D rendering.
Virtual try-on systems allow users to interactively try different products within VR scenarios. However, most existing VTON methods operate only on predefined eyewear templates and lack support for fine-grained, user-driven customization. While GlassesGAN enables personalized 2D eyewear design, its capability remains limited to 2D image generation. Motivated by the success of 3D Gaussian Blendshapes in head reconstruction, we integrate these two techniques and propose GlassesGB, a framework that supports customizable eyewear generation for 3D head avatars. GlassesGB effectively bridges 2D generative customization with 3D head avatar rendering, addressing the challenge in achieving personalized eyewear design for VR applications. The implementation code is available at https://ruiyangju.github.io/GlassesGB.