Deep learning based super-resolution achieves high-quality results, but its heavy computational workloads, large buffer, and high external memory bandwidth inhibit its usage in mobile devices. To solve the above issues, this paper proposes a real-time hardware accelerator with the tilted layer fusion method that reduces external DRAM bandwidth by 92% and just needs 102KB on-chip memory. The design implemented with a 40nm CMOS process achieves 1920x1080@60fps throughput with 544.3K gate count when running at 600MHz; it has higher throughput and lower area cost than previous designs.