Recent measurement devices for the Bidirectional Texture Function (BTF) offer high parallel capture of the spatial varying reflectance data for complex material samples like metallic and pearlescent car paint. We extend the hybrid model and image based car paint rendering approach from Rump et al. [15]. We use measurement data from homogenous materials or materials with stochastical mesostructure to obtain data samples for the fitting of BRDF models like Cook-Torrance. We applied this approach to measured samples of metallic car paint and use the BRDF model from Rump et al. to enable simple editing of the car paint's color by dividing the BTF into a model-based part, a direction-dependent color table and the spatial varying glitter effects from the effect flakes in modern car paints.