360-degree video coding using geometry projection

An apparatus comprising:

• a processor configured to:
• obtain a current sample location in a frame-packed picture, wherein the frame-packed picture comprises a plurality of faces associated with a first projection geometry, wherein the current sample location is located in a first face associated with the frame-packed picture, wherein the current sample location is associated with a current sample of a video content;
• obtain a first prediction sample location, wherein the current sample is to be predicted based on the first prediction sample location;
• determine that the first prediction sample location is outside of a boundary of the first face;
• based on the determination that the first prediction sample location is outside of the first face, obtain a 3D position of the first prediction sample location, wherein the 3D position of the first prediction sample location is associated with the first face;
• based on the 3D position of the first prediction sample location, obtain a second face that includes a second prediction sample location;
• apply a geometry projection with the 3D position of the first prediction sample location to derive a 2D planar position of the first prediction sample location in the second face, wherein the 2D planar position of the first prediction sample location is the second prediction sample location;
• and
• predict the current sample associated with the video content based on the second prediction sample location.