Consolidating Attention Features for Multi-view Image Editing