diff --git a/src/Math/Math.h b/src/Math/Math.h
index 7b56d7f5a..a0217f88c 100644
--- a/src/Math/Math.h
+++ b/src/Math/Math.h
@@ -16,6 +16,8 @@
 */
 
 #include <cstddef>
+#include <type_traits>
+#include <limits>
 
 #include "magnumVisibility.h"
 
@@ -85,6 +87,43 @@ template<size_t exponent, class T> inline constexpr T pow(T base) {
  */
 size_t MAGNUM_EXPORT log(size_t base, size_t number);
 
+/**
+@brief Normalize floating-point value
+
+Converts integral value from full range of given (signed/unsigned) integral
+type to value in range @f$ [0, 1] @f$.
+
+@attention To ensure the integral type is correctly detected when using
+literals, this function should be called with both template parameters
+explicit, e.g.:
+@code
+// Even if this is char literal, integral type is `int`, thus a = 0.1f
+float a = normalize<float>('\127');
+
+// b = 1.0f
+float b = normalize<float, char>('\127');
+@endcode
+*/
+template<class FloatingPoint, class Integral> inline constexpr typename std::enable_if<std::is_floating_point<FloatingPoint>::value && std::is_integral<Integral>::value, FloatingPoint>::type normalize(Integral value) {
+    return (FloatingPoint(value)-FloatingPoint(std::numeric_limits<Integral>::min()))/
+        (FloatingPoint(std::numeric_limits<Integral>::max()) - FloatingPoint(std::numeric_limits<Integral>::min()));
+}
+
+/**
+@brief Denormalize floating-point value
+
+Converts floating-point value in range @f$ [0, 1] @f$ to full range of given
+integral type.
+
+@note For best precision, `FloatingPoint` type should be always larger that
+resulting `Integral` type (e.g. `double` to `int`, `long double` to `long long`).
+*/
+template<class Integral, class FloatingPoint> inline constexpr typename std::enable_if<std::is_floating_point<FloatingPoint>::value && std::is_integral<Integral>::value, Integral>::type denormalize(FloatingPoint value) {
+    return             std::numeric_limits<Integral>::min() +
+        Integral(value*std::numeric_limits<Integral>::max()) -
+        Integral(value*std::numeric_limits<Integral>::min());
+}
+
 /**
  * @brief Angle in degrees
  *
diff --git a/src/Math/Test/MathTest.cpp b/src/Math/Test/MathTest.cpp
index 46babe120..dc02165f5 100644
--- a/src/Math/Test/MathTest.cpp
+++ b/src/Math/Test/MathTest.cpp
@@ -17,12 +17,16 @@
 
 #include "Math.h"
 
+using namespace std;
+
 CORRADE_TEST_MAIN(Magnum::Math::Test::MathTest)
 
 namespace Magnum { namespace Math { namespace Test {
 
 MathTest::MathTest() {
     addTests(&MathTest::degrad,
+             &MathTest::normalize,
+             &MathTest::denormalize,
              &MathTest::pow,
              &MathTest::log);
 }
@@ -33,6 +37,52 @@ void MathTest::degrad() {
     CORRADE_COMPARE(rad(Constants<double>::pi()/2), Constants<double>::pi()/2);
 }
 
+void MathTest::normalize() {
+    /* Range for signed and unsigned */
+    CORRADE_COMPARE((Math::normalize<float, char>(-128)), 0.0f);
+    CORRADE_COMPARE((Math::normalize<float, char>(127)), 1.0f);
+    CORRADE_COMPARE((Math::normalize<float, unsigned char>(0)), 0.0f);
+    CORRADE_COMPARE((Math::normalize<float, unsigned char>(255)), 1.0f);
+
+    /* Between */
+    CORRADE_COMPARE((Math::normalize<float, short>(16384)), 0.750011f);
+    CORRADE_COMPARE((Math::normalize<float, short>(-16384)), 0.250004f);
+
+    /* Test overflow for large types */
+    CORRADE_COMPARE((Math::normalize<float, int>(numeric_limits<int>::min())), 0.0f);
+    CORRADE_COMPARE((Math::normalize<float, int>(numeric_limits<int>::max())), 1.0f);
+    CORRADE_COMPARE((Math::normalize<float, unsigned int>(0)), 0.0f);
+    CORRADE_COMPARE((Math::normalize<float, unsigned int>(numeric_limits<unsigned int>::max())), 1.0f);
+
+    CORRADE_COMPARE((Math::normalize<double, long long>(numeric_limits<long long>::min())), 0.0);
+    CORRADE_COMPARE((Math::normalize<double, long long>(numeric_limits<long long>::max())), 1.0);
+    CORRADE_COMPARE((Math::normalize<double, unsigned long long>(0)), 0.0);
+    CORRADE_COMPARE((Math::normalize<double, unsigned long long>(numeric_limits<unsigned long long>::max())), 1.0);
+}
+
+void MathTest::denormalize() {
+    /* Range for signed and unsigned */
+    CORRADE_COMPARE(Math::denormalize<char>(0.0f), -128);
+    CORRADE_COMPARE(Math::denormalize<char>(1.0f), 127);
+    CORRADE_COMPARE(Math::denormalize<unsigned char>(0.0f), 0);
+    CORRADE_COMPARE(Math::denormalize<unsigned char>(1.0f), 255);
+
+    /* Between */
+    CORRADE_COMPARE(Math::denormalize<short>(0.33f), -11142);
+    CORRADE_COMPARE(Math::denormalize<short>(0.66f), 10484);
+
+    /* Test overflow for large types */
+    CORRADE_COMPARE(Math::denormalize<int>(0.0f), numeric_limits<int>::min());
+    CORRADE_COMPARE(Math::denormalize<unsigned int>(0.0f), 0);
+    CORRADE_COMPARE(Math::denormalize<long long>(0.0), numeric_limits<long long>::min());
+    CORRADE_COMPARE(Math::denormalize<unsigned long long>(0.0), 0);
+
+    CORRADE_COMPARE(Math::denormalize<int>(1.0), numeric_limits<int>::max());
+    CORRADE_COMPARE(Math::denormalize<unsigned int>(1.0), numeric_limits<unsigned int>::max());
+    CORRADE_COMPARE((Math::denormalize<long long, long double>(1.0)), numeric_limits<long long>::max());
+    CORRADE_COMPARE((Math::denormalize<unsigned long long, long double>(1.0)), numeric_limits<unsigned long long>::max());
+}
+
 void MathTest::pow() {
     CORRADE_COMPARE(Math::pow<10>(2ul), 1024ul);
     CORRADE_COMPARE(Math::pow<0>(3ul), 1ul);
diff --git a/src/Math/Test/MathTest.h b/src/Math/Test/MathTest.h
index 548b01e3c..b6dcba0c8 100644
--- a/src/Math/Test/MathTest.h
+++ b/src/Math/Test/MathTest.h
@@ -24,6 +24,8 @@ class MathTest: public Corrade::TestSuite::Tester<MathTest> {
         MathTest();
 
         void degrad();
+        void normalize();
+        void denormalize();
         void pow();
         void log();
 };