diff --git a/tests/validation/CL/AbsoluteDifference.cpp b/tests/validation/CL/AbsoluteDifference.cpp
new file mode 100644
index 0000000..cabb76f
--- /dev/null
+++ b/tests/validation/CL/AbsoluteDifference.cpp
@@ -0,0 +1,137 @@
+/*
+ * Copyright (c) 2017 ARM Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include "arm_compute/core/Types.h"
+#include "arm_compute/runtime/CL/CLTensor.h"
+#include "arm_compute/runtime/CL/CLTensorAllocator.h"
+#include "arm_compute/runtime/CL/functions/CLAbsoluteDifference.h"
+#include "tests/CL/CLAccessor.h"
+#include "tests/PaddingCalculator.h"
+#include "tests/datasets/ConvertPolicyDataset.h"
+#include "tests/datasets/ShapeDatasets.h"
+#include "tests/framework/Asserts.h"
+#include "tests/framework/Macros.h"
+#include "tests/framework/datasets/Datasets.h"
+#include "tests/validation/Validation.h"
+#include "tests/validation/fixtures/AbsoluteDifferenceFixture.h"
+
+namespace arm_compute
+{
+namespace test
+{
+namespace validation
+{
+namespace
+{
+/** Input data sets **/
+const auto AbsoluteDifferenceU8Dataset = combine(combine(framework::dataset::make("DataType", DataType::U8), framework::dataset::make("DataType", DataType::U8)), framework::dataset::make("DataType",
+                                                 DataType::U8));
+const auto AbsoluteDifferenceS16Dataset = combine(combine(framework::dataset::make("DataType", { DataType::U8, DataType::S16 }), framework::dataset::make("DataType", DataType::S16)),
+                                                  framework::dataset::make("DataType", DataType::S16));
+} // namespace
+
+TEST_SUITE(CL)
+TEST_SUITE(AbsoluteDifference)
+
+template <typename T>
+using CLAbsoluteDifferenceFixture = AbsoluteDifferenceValidationFixture<CLTensor, CLAccessor, CLAbsoluteDifference, T>;
+
+TEST_SUITE(U8)
+DATA_TEST_CASE(Configuration, framework::DatasetMode::ALL, combine(concat(datasets::SmallShapes(), datasets::LargeShapes()), AbsoluteDifferenceU8Dataset),
+               shape, data_type0, data_type1, output_data_type)
+{
+    // Create tensors
+    CLTensor ref_src1 = create_tensor<CLTensor>(shape, data_type0);
+    CLTensor ref_src2 = create_tensor<CLTensor>(shape, data_type1);
+    CLTensor dst      = create_tensor<CLTensor>(shape, output_data_type);
+
+    // Create and Configure function
+    CLAbsoluteDifference abs_diff;
+    abs_diff.configure(&ref_src1, &ref_src2, &dst);
+
+    // Validate valid region
+    const ValidRegion valid_region = shape_to_valid_region(shape);
+    validate(dst.info()->valid_region(), valid_region);
+
+    // Validate padding
+    const PaddingSize padding = PaddingCalculator(shape.x(), 16).required_padding();
+    validate(ref_src1.info()->padding(), padding);
+    validate(ref_src2.info()->padding(), padding);
+    validate(dst.info()->padding(), padding);
+}
+
+FIXTURE_DATA_TEST_CASE(RunSmall, CLAbsoluteDifferenceFixture<uint8_t>, framework::DatasetMode::PRECOMMIT, combine(datasets::SmallShapes(), AbsoluteDifferenceU8Dataset))
+{
+    // Validate output
+    validate(CLAccessor(_target), _reference);
+}
+FIXTURE_DATA_TEST_CASE(RunLarge, CLAbsoluteDifferenceFixture<uint8_t>, framework::DatasetMode::NIGHTLY, combine(datasets::LargeShapes(), AbsoluteDifferenceU8Dataset))
+{
+    // Validate output
+    validate(CLAccessor(_target), _reference);
+}
+TEST_SUITE_END()
+
+TEST_SUITE(S16)
+DATA_TEST_CASE(Configuration, framework::DatasetMode::ALL, combine(concat(datasets::SmallShapes(), datasets::LargeShapes()), AbsoluteDifferenceS16Dataset),
+               shape, data_type0, data_type1, output_data_type)
+{
+    // Create tensors
+    CLTensor ref_src1 = create_tensor<CLTensor>(shape, data_type0);
+    CLTensor ref_src2 = create_tensor<CLTensor>(shape, data_type1);
+    CLTensor dst      = create_tensor<CLTensor>(shape, output_data_type);
+
+    // Create and Configure function
+    CLAbsoluteDifference abs_diff;
+    abs_diff.configure(&ref_src1, &ref_src2, &dst);
+
+    // Validate valid region
+    const ValidRegion valid_region = shape_to_valid_region(shape);
+    validate(dst.info()->valid_region(), valid_region);
+
+    // Validate padding
+    const PaddingSize padding = PaddingCalculator(shape.x(), 16).required_padding();
+    validate(ref_src1.info()->padding(), padding);
+    validate(ref_src2.info()->padding(), padding);
+    validate(dst.info()->padding(), padding);
+}
+
+FIXTURE_DATA_TEST_CASE(RunSmall, CLAbsoluteDifferenceFixture<int16_t>, framework::DatasetMode::PRECOMMIT, combine(datasets::SmallShapes(), AbsoluteDifferenceS16Dataset))
+{
+    // Validate output
+    validate(CLAccessor(_target), _reference);
+}
+
+FIXTURE_DATA_TEST_CASE(RunLarge, CLAbsoluteDifferenceFixture<int16_t>, framework::DatasetMode::NIGHTLY, combine(datasets::LargeShapes(), AbsoluteDifferenceS16Dataset))
+{
+    // Validate output
+    validate(CLAccessor(_target), _reference);
+}
+TEST_SUITE_END()
+
+TEST_SUITE_END()
+TEST_SUITE_END()
+
+} // namespace validation
+} // namespace test
+} // namespace arm_compute
diff --git a/tests/validation/CPP/AbsoluteDifference.cpp b/tests/validation/CPP/AbsoluteDifference.cpp
new file mode 100644
index 0000000..f518e67
--- /dev/null
+++ b/tests/validation/CPP/AbsoluteDifference.cpp
@@ -0,0 +1,60 @@
+/*
+ * Copyright (c) 2017 ARM Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include "AbsoluteDifference.h"
+
+#include "arm_compute/core/Types.h"
+#include "tests/validation/FixedPoint.h"
+#include "tests/validation/Helpers.h"
+
+namespace arm_compute
+{
+namespace test
+{
+namespace validation
+{
+namespace reference
+{
+template <typename T>
+SimpleTensor<T> absolute_difference(const SimpleTensor<T> &src1, const SimpleTensor<T> &src2, DataType dst_data_type)
+{
+    SimpleTensor<T> result(src1.shape(), dst_data_type);
+
+    using intermediate_type = typename common_promoted_signed_type<T>::intermediate_type;
+
+    for(int i = 0; i < src1.num_elements(); ++i)
+    {
+        intermediate_type val = std::abs(static_cast<intermediate_type>(src1[i]) - static_cast<intermediate_type>(src2[i]));
+        result[i]             = saturate_cast<T>(val);
+    }
+
+    return result;
+}
+
+template SimpleTensor<uint8_t> absolute_difference(const SimpleTensor<uint8_t> &src1, const SimpleTensor<uint8_t> &src2, DataType dst_data_type);
+template SimpleTensor<int16_t> absolute_difference(const SimpleTensor<int16_t> &src1, const SimpleTensor<int16_t> &src2, DataType dst_data_type);
+template SimpleTensor<int8_t> absolute_difference(const SimpleTensor<int8_t> &src1, const SimpleTensor<int8_t> &src2, DataType dst_data_type);
+} // namespace reference
+} // namespace validation
+} // namespace test
+} // namespace arm_compute
diff --git a/tests/validation/CPP/AbsoluteDifference.h b/tests/validation/CPP/AbsoluteDifference.h
new file mode 100644
index 0000000..b14da15
--- /dev/null
+++ b/tests/validation/CPP/AbsoluteDifference.h
@@ -0,0 +1,44 @@
+/*
+ * Copyright (c) 2017 ARM Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef __ARM_COMPUTE_TEST_ABSOLUTE_DIFFERENCE_H__
+#define __ARM_COMPUTE_TEST_ABSOLUTE_DIFFERENCE_H__
+
+#include "tests/SimpleTensor.h"
+#include "tests/validation/Helpers.h"
+
+namespace arm_compute
+{
+namespace test
+{
+namespace validation
+{
+namespace reference
+{
+template <typename T>
+SimpleTensor<T> absolute_difference(const SimpleTensor<T> &src1, const SimpleTensor<T> &src2, DataType dst_data_type);
+} // namespace reference
+} // namespace validation
+} // namespace test
+} // namespace arm_compute
+#endif /* __ARM_COMPUTE_TEST_ABSOLUTE_DIFFERENCE_H__ */
diff --git a/tests/validation/NEON/AbsoluteDifference.cpp b/tests/validation/NEON/AbsoluteDifference.cpp
new file mode 100644
index 0000000..74c5701
--- /dev/null
+++ b/tests/validation/NEON/AbsoluteDifference.cpp
@@ -0,0 +1,136 @@
+/*
+ * Copyright (c) 2017 ARM Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONCLCTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include "arm_compute/core/Types.h"
+#include "arm_compute/runtime/NEON/functions/NEAbsoluteDifference.h"
+#include "arm_compute/runtime/Tensor.h"
+#include "arm_compute/runtime/TensorAllocator.h"
+#include "tests/NEON/Accessor.h"
+#include "tests/PaddingCalculator.h"
+#include "tests/datasets/ConvertPolicyDataset.h"
+#include "tests/datasets/ShapeDatasets.h"
+#include "tests/framework/Asserts.h"
+#include "tests/framework/Macros.h"
+#include "tests/framework/datasets/Datasets.h"
+#include "tests/validation/Validation.h"
+#include "tests/validation/fixtures/AbsoluteDifferenceFixture.h"
+
+namespace arm_compute
+{
+namespace test
+{
+namespace validation
+{
+namespace
+{
+/** Input data sets **/
+const auto AbsoluteDifferenceU8Dataset = combine(combine(framework::dataset::make("DataType", DataType::U8), framework::dataset::make("DataType", DataType::U8)), framework::dataset::make("DataType",
+                                                 DataType::U8));
+const auto AbsoluteDifferenceS16Dataset = combine(combine(framework::dataset::make("DataType", { DataType::U8, DataType::S16 }), framework::dataset::make("DataType", DataType::S16)),
+                                                  framework::dataset::make("DataType", DataType::S16));
+} // namespace
+
+TEST_SUITE(NEON)
+TEST_SUITE(AbsoluteDifference)
+
+template <typename T>
+using NEAbsoluteDifferenceFixture = AbsoluteDifferenceValidationFixture<Tensor, Accessor, NEAbsoluteDifference, T>;
+
+TEST_SUITE(U8)
+DATA_TEST_CASE(Configuration, framework::DatasetMode::ALL, combine(concat(datasets::SmallShapes(), datasets::LargeShapes()), AbsoluteDifferenceU8Dataset),
+               shape, data_type0, data_type1, output_data_type)
+{
+    // Create tensors
+    Tensor ref_src1 = create_tensor<Tensor>(shape, data_type0);
+    Tensor ref_src2 = create_tensor<Tensor>(shape, data_type1);
+    Tensor dst      = create_tensor<Tensor>(shape, output_data_type);
+
+    // Create and Configure function
+    NEAbsoluteDifference abs_diff;
+    abs_diff.configure(&ref_src1, &ref_src2, &dst);
+
+    // Validate valid region
+    const ValidRegion valid_region = shape_to_valid_region(shape);
+    validate(dst.info()->valid_region(), valid_region);
+
+    // Validate padding
+    const PaddingSize padding = PaddingCalculator(shape.x(), 16).required_padding();
+    validate(ref_src1.info()->padding(), padding);
+    validate(ref_src2.info()->padding(), padding);
+    validate(dst.info()->padding(), padding);
+}
+
+FIXTURE_DATA_TEST_CASE(RunSmall, NEAbsoluteDifferenceFixture<uint8_t>, framework::DatasetMode::PRECOMMIT, combine(datasets::SmallShapes(), AbsoluteDifferenceU8Dataset))
+{
+    // Validate output
+    validate(Accessor(_target), _reference);
+}
+FIXTURE_DATA_TEST_CASE(RunLarge, NEAbsoluteDifferenceFixture<uint8_t>, framework::DatasetMode::NIGHTLY, combine(datasets::LargeShapes(), AbsoluteDifferenceU8Dataset))
+{
+    // Validate output
+    validate(Accessor(_target), _reference);
+}
+TEST_SUITE_END()
+
+TEST_SUITE(S16)
+DATA_TEST_CASE(Configuration, framework::DatasetMode::ALL, combine(framework::dataset::concat(datasets::SmallShapes(), datasets::LargeShapes()), AbsoluteDifferenceS16Dataset),
+               shape, data_type0, data_type1, output_data_type)
+{
+    // Create tensors
+    Tensor ref_src1 = create_tensor<Tensor>(shape, data_type0);
+    Tensor ref_src2 = create_tensor<Tensor>(shape, data_type1);
+    Tensor dst      = create_tensor<Tensor>(shape, output_data_type);
+
+    // Create and Configure function
+    NEAbsoluteDifference abs_diff;
+    abs_diff.configure(&ref_src1, &ref_src2, &dst);
+
+    // Validate valid region
+    const ValidRegion valid_region = shape_to_valid_region(shape);
+    validate(dst.info()->valid_region(), valid_region);
+
+    // Validate padding
+    const PaddingSize padding = PaddingCalculator(shape.x(), 16).required_padding();
+    validate(ref_src1.info()->padding(), padding);
+    validate(ref_src2.info()->padding(), padding);
+    validate(dst.info()->padding(), padding);
+}
+
+FIXTURE_DATA_TEST_CASE(RunSmall, NEAbsoluteDifferenceFixture<int16_t>, framework::DatasetMode::PRECOMMIT, combine(datasets::SmallShapes(), AbsoluteDifferenceS16Dataset))
+{
+    // Validate output
+    validate(Accessor(_target), _reference);
+}
+
+FIXTURE_DATA_TEST_CASE(RunLarge, NEAbsoluteDifferenceFixture<int16_t>, framework::DatasetMode::NIGHTLY, combine(datasets::LargeShapes(), AbsoluteDifferenceS16Dataset))
+{
+    // Validate output
+    validate(Accessor(_target), _reference);
+}
+TEST_SUITE_END()
+
+TEST_SUITE_END()
+TEST_SUITE_END()
+} // namespace validation
+} // namespace test
+} // namespace arm_compute
diff --git a/tests/validation/fixtures/AbsoluteDifferenceFixture.h b/tests/validation/fixtures/AbsoluteDifferenceFixture.h
new file mode 100644
index 0000000..54e202d
--- /dev/null
+++ b/tests/validation/fixtures/AbsoluteDifferenceFixture.h
@@ -0,0 +1,114 @@
+/*
+ * Copyright (c) 2017 ARM Limited.
+ *
+ * SPDX-License-Identifier: MIT
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef ARM_COMPUTE_TEST_ABSOLUTE_DIFFERENCE_FIXTURE
+#define ARM_COMPUTE_TEST_ABSOLUTE_DIFFERENCE_FIXTURE
+
+#include "arm_compute/core/TensorShape.h"
+#include "arm_compute/core/Types.h"
+#include "tests/AssetsLibrary.h"
+#include "tests/Globals.h"
+#include "tests/IAccessor.h"
+#include "tests/framework/Asserts.h"
+#include "tests/framework/Fixture.h"
+#include "tests/validation/CPP/AbsoluteDifference.h"
+#include "tests/validation/Helpers.h"
+
+namespace arm_compute
+{
+namespace test
+{
+namespace validation
+{
+template <typename TensorType, typename AccessorType, typename FunctionType, typename T>
+class AbsoluteDifferenceValidationFixture : public framework::Fixture
+{
+public:
+    template <typename...>
+    void setup(TensorShape shape, DataType data_type0, DataType data_type1, DataType output_data_type)
+    {
+        _target    = compute_target(shape, data_type0, data_type1, output_data_type);
+        _reference = compute_reference(shape, data_type0, data_type1, output_data_type);
+    }
+
+protected:
+    template <typename U>
+    void fill(U &&tensor, int i)
+    {
+        library->fill_tensor_uniform(tensor, i);
+    }
+
+    TensorType compute_target(const TensorShape &shape, DataType data_type0, DataType data_type1, DataType output_data_type)
+    {
+        // Create tensors
+        TensorType ref_src1 = create_tensor<TensorType>(shape, data_type0, 1);
+        TensorType ref_src2 = create_tensor<TensorType>(shape, data_type1, 1);
+        TensorType dst      = create_tensor<TensorType>(shape, output_data_type, 1);
+
+        // Create and configure function
+        FunctionType abs_diff;
+        abs_diff.configure(&ref_src1, &ref_src2, &dst);
+
+        ARM_COMPUTE_EXPECT(ref_src1.info()->is_resizable(), framework::LogLevel::ERRORS);
+        ARM_COMPUTE_EXPECT(ref_src2.info()->is_resizable(), framework::LogLevel::ERRORS);
+        ARM_COMPUTE_EXPECT(dst.info()->is_resizable(), framework::LogLevel::ERRORS);
+
+        // Allocate tensors
+        ref_src1.allocator()->allocate();
+        ref_src2.allocator()->allocate();
+        dst.allocator()->allocate();
+
+        ARM_COMPUTE_EXPECT(!ref_src1.info()->is_resizable(), framework::LogLevel::ERRORS);
+        ARM_COMPUTE_EXPECT(!ref_src2.info()->is_resizable(), framework::LogLevel::ERRORS);
+        ARM_COMPUTE_EXPECT(!dst.info()->is_resizable(), framework::LogLevel::ERRORS);
+
+        // Fill tensors
+        fill(AccessorType(ref_src1), 0);
+        fill(AccessorType(ref_src2), 1);
+
+        // Compute function
+        abs_diff.run();
+
+        return dst;
+    }
+
+    SimpleTensor<T> compute_reference(const TensorShape &shape, DataType data_type0, DataType data_type1, DataType output_data_type)
+    {
+        // Create reference
+        SimpleTensor<T> ref_src1{ shape, data_type0, 1 };
+        SimpleTensor<T> ref_src2{ shape, data_type1, 1 };
+
+        // Fill reference
+        fill(ref_src1, 0);
+        fill(ref_src2, 1);
+
+        return reference::absolute_difference<T>(ref_src1, ref_src2, output_data_type);
+    }
+
+    TensorType      _target{};
+    SimpleTensor<T> _reference{};
+};
+} // namespace validation
+} // namespace test
+} // namespace arm_compute
+#endif /* ARM_COMPUTE_TEST_ABSOLUTE_DIFFERENCE_FIXTURE */