Cycles Denoising: Add more robust outlier heuristic to avoid artifacts

Extremely bright pixels in the rendered image cause the denoising algorithm
to produce extremely noticable artifacts. Therefore, a heuristic is needed
to exclude these pixels from the filtering process.

The new approach calculates the 75% percentile of the 5x5 neighborhood of
each pixel and flags the pixel if it is more than twice as bright.

During the reconstruction process, flagged pixels are skipped. Therefore,
they don't cause any problems for neighboring pixels, and the outlier pixels
themselves are replaced by a prediction of their actual value based on their
feature pass values and the neighboring pixels.

Therefore, the denoiser now also works as a smarter despeckling filter that
uses a more accurate prediction of the pixel instead of a simple average.
This can be used even if denoising isn't wanted by setting the denoising
radius to 1.

1 files changed, 16 insertions, 0 deletions

diff --git a/intern/cycles/kernel/kernels/cuda/filter.cu b/intern/cycles/kernel/kernels/cuda/filter.cu
index 50f73f9728d..f812a6601c6 100644
--- a/intern/cycles/kernel/kernels/cuda/filter.cu
+++ b/intern/cycles/kernel/kernels/cuda/filter.cu
@@ -88,6 +88,22 @@ kernel_cuda_filter_get_feature(int sample,
 
 extern "C" __global__ void
 CUDA_LAUNCH_BOUNDS(CUDA_THREADS_BLOCK_WIDTH, CUDA_KERNEL_MAX_REGISTERS)
+kernel_cuda_filter_detect_outliers(float *image,
+                                   float *variance,
+                                   float *depth,
+                                   float *output,
+                                   int4 prefilter_rect,
+                                   int pass_stride)
+{
+	int x = prefilter_rect.x + blockDim.x*blockIdx.x + threadIdx.x;
+	int y = prefilter_rect.y + blockDim.y*blockIdx.y + threadIdx.y;
+	if(x < prefilter_rect.z && y < prefilter_rect.w) {
+		kernel_filter_detect_outliers(x, y, image, variance, depth, output, prefilter_rect, pass_stride);
+	}
+}
+
+extern "C" __global__ void
+CUDA_LAUNCH_BOUNDS(CUDA_THREADS_BLOCK_WIDTH, CUDA_KERNEL_MAX_REGISTERS)
 kernel_cuda_filter_combine_halves(float *mean, float *variance, float *a, float *b, int4 prefilter_rect, int r)
 {
 	int x = prefilter_rect.x + blockDim.x*blockIdx.x + threadIdx.x;