Shanghai Normal University

At LEDNet's existing dynamic-kernel interfaces, LEDNet-P3 adds a residual branch that splits the kernel increment into a signed kernel-sum coordinate and four zero-sum shape bases, with the degraded RGB image feeding only the shape coefficients. On 1,800 official LOL-Blur test pairs, the paired LEDNet comparison shows structural similarity up 0.004656 and learned perceptual patch similarity down 0.011445, while the PSNR difference is −0.045441 dB, so a PSNR improvement is not established. The branch adds 226,880 parameters and about 17.7% measured forward latency.

At LEDNet's existing dynamic-kernel interfaces, LEDNet-P3 adds a residual branch that splits the kernel increment into a signed kernel-sum coordinate and four zero-sum shape bases, with the degraded RGB image feeding only the shape coefficients. On 1,800 official LOL-Blur test pairs, the paired LEDNet comparison shows structural similarity up 0.004656 and learned perceptual patch similarity down 0.011445, while the PSNR difference is −0.045441 dB, so a PSNR improvement is not established. The branch adds 226,880 parameters and about 17.7% measured forward latency.