Using the first reference image as the EXACT scene, location, lighting, camera angle and background — keep everything identical (same marble surface, same window light, same blurred interior, same perspective and shadows).
Replace the small folded capsule toy with a fully unfolded, assembled 3D miniature model of the building shown in the SECOND reference image. The toy has "popped open" after the metal button was pressed and transformed into the complete, detailed building standing on the same spot.
The model keeps the same collectible toy aesthetic: detailed sculpted ceramic / resin material, fine engraved patterns, subtle gold trim lines, authentic architecture, accurate proportions, domes, towers, arches, windows, ornaments and the real color palette of the building.
The building sits exactly where the capsule toy was, same size footprint, casting realistic soft shadows and reflections on the marble, perfectly integrated into the original scene.
Style: ultra-realistic product photography, 3D render, high detail, shallow depth of field, cinematic natural daylight, 9:16 vertical format.
