Build memory-efficient sparse voxel octrees with GPU-accelerated mesh voxelization using OpenCL and bit-packed storage.