Color vision deficiency (CVD) affects a significant portion of the population and poses persistent challenges to visual content perception, particularly in immersive 3D environments. While existing color enhancement methods primarily focus on 2D images, they lack explicit 3D scene representations and fail to ensure color consistency across viewpoints. This limitation reduces their applicability to 3D-aware tasks, such as the digital visualization of complex textile patterns with fine-grained color variations. In this paper, we present a fast, CVD-friendly 3D content generation framework that integrates perceptual color enhancement with efficient multi-view 3D reconstruction. Our approach first applies a conditional GAN to enhance color discriminability for CVD observers on a per-view basis. These enhanced views are then integrated across viewpoints and used as supervision to reconstruct a colorblind-friendly 3D scene using 3D Gaussian Splatting. A multi-view color consistency constraint is introduced to suppress view-dependent color drift. Importantly, the proposed method preserves the original Gaussian optimization pipeline while achieving perceptually enhanced rendering without compromising reconstruction efficiency. This framework ensures color discriminability and spatial consistency, making it suitable for accessibility-oriented applications in 3D textile display.

color vision deficiency, 3D gaussian splatting, multi-view 3D reconstruction, digital textiles