swscale/filters: hard-code radius for trivial kernels

box() and triangle() have well-defined, trivially verifiable numerical
inverses.

We could actually pre-compute and hard-code the numerical inverse of all
non-parametric kernels, but I'm a bit reluctant to do this as I have plans to
adjust the value of SWS_MAX_REDUCE_CUTOFF based on the desired bit depth of the
output, which makes a hard-coding approach unfeasible.

(It would also be a brittle solution that may break whenever we extend the
scaler configuration API, as well as making it harder to add new filters)

Signed-off-by: Niklas Haas <git@haasn.dev>

libswscale/filters.c

@@ -154,7 +154,11 @@ static bool validate_params(const SwsFilterFunction *fun, SwsScaler scaler)
     }
 }
 
-static double filter_radius(const SwsFilterFunction *fun)
+/**
+ * Numerically estimate the last intersection between the function value
+ * and the cutoff domain [-SWS_MAX_REDUCE_CUTOFF, SWS_MAX_REDUCE_CUTOFF].
+ */
+static double est_filter_radius(const SwsFilterFunction *fun)
 {
     const double bound = fun->radius;
     const double step  = 1e-2;
@@ -225,7 +229,21 @@ int ff_sws_filter_generate(void *log, const SwsFilterParams *params,
     if (fun.radius < 0.0) /* tunable width kernels like lanczos */
         fun.radius = fun.params[0];
 
-    const double radius = filter_radius(&fun) * stretch;
+    double radius;
+    switch (scaler) {
+    case SWS_SCALE_POINT:
+        radius = 0.5;
+        break;
+    case SWS_SCALE_BILINEAR:
+        radius = 1.0 - SWS_MAX_REDUCE_CUTOFF;
+        break;
+    default:
+        /* Numerically estimate radius of nontrivial or parametric kernels */
+        radius = est_filter_radius(&fun);
+        break;
+    }
+    radius *= stretch;
+
     int filter_size = ceil(radius * 2.0);
     filter_size = FFMIN(filter_size, params->src_size);
     av_assert0(filter_size >= 1);