diff --git a/doc/source/whatsnew/v0.23.0.txt b/doc/source/whatsnew/v0.23.0.txt
index 72f63a4da0f4d..c9c3c899bce61 100644
--- a/doc/source/whatsnew/v0.23.0.txt
+++ b/doc/source/whatsnew/v0.23.0.txt
@@ -823,6 +823,7 @@ Sparse
 - Bug in :class:`SparseDataFrame.to_csv` causing exception (:issue:`19384`)
 - Bug in :class:`SparseSeries.memory_usage` which caused segfault by accessing non sparse elements (:issue:`19368`)
 - Bug in constructing a ``SparseArray``: if ``data`` is a scalar and ``index`` is defined it will coerce to ``float64`` regardless of scalar's dtype. (:issue:`19163`)
+- Bug in :func:`SparseSeries.unique` which returns only sparse elements during unique (:issue:`19651`)
 
 Reshaping
 ^^^^^^^^^
diff --git a/pandas/core/algorithms.py b/pandas/core/algorithms.py
index c754c063fce8e..51355db9f7d8a 100644
--- a/pandas/core/algorithms.py
+++ b/pandas/core/algorithms.py
@@ -10,7 +10,8 @@
     maybe_promote, construct_1d_object_array_from_listlike)
 from pandas.core.dtypes.generic import (
     ABCSeries, ABCIndex,
-    ABCIndexClass, ABCCategorical)
+    ABCIndexClass, ABCCategorical,
+    ABCSparseArray)
 from pandas.core.dtypes.common import (
     is_unsigned_integer_dtype, is_signed_integer_dtype,
     is_integer_dtype, is_complex_dtype,
@@ -362,7 +363,14 @@ def unique(values):
     htable, _, values, dtype, ndtype = _get_hashtable_algo(values)
 
     table = htable(len(values))
-    uniques = table.unique(values)
+
+    if isinstance(values, ABCSparseArray):
+        to_unique = values.sp_values
+        if values.sp_index.ngaps > 0:
+            to_unique = np.append(to_unique, [values.fill_value])
+        uniques = table.unique(to_unique)
+    else:
+        uniques = table.unique(values)
     uniques = _reconstruct_data(uniques, dtype, original)
 
     if isinstance(original, ABCSeries) and is_datetime64tz_dtype(dtype):
@@ -469,6 +477,7 @@ def factorize(values, sort=False, order=None, na_sentinel=-1, size_hint=None):
     table = hash_klass(size_hint or len(values))
     uniques = vec_klass()
     check_nulls = not is_integer_dtype(original)
+
     labels = table.get_labels(values, uniques, 0, na_sentinel, check_nulls)
 
     labels = _ensure_platform_int(labels)
diff --git a/pandas/tests/test_algos.py b/pandas/tests/test_algos.py
index b1e3177547ac6..bc3eb74502c6d 100644
--- a/pandas/tests/test_algos.py
+++ b/pandas/tests/test_algos.py
@@ -8,7 +8,8 @@
 from datetime import datetime
 from itertools import permutations
 from pandas import (Series, Categorical, CategoricalIndex,
-                    Timestamp, DatetimeIndex, Index, IntervalIndex)
+                    Timestamp, DatetimeIndex, Index, IntervalIndex,
+                    SparseArray)
 import pandas as pd
 
 from pandas import compat
@@ -268,6 +269,16 @@ def test_object_refcount_bug(self):
         for i in range(1000):
             len(algos.unique(lst))
 
+    @pytest.mark.parametrize('fill_value', [0, 1, np.nan, None])
+    def test_sparse(self, fill_value):
+        # GH 19595
+        arr = SparseArray([0, 1, np.nan, None], fill_value=fill_value)
+
+        result = algos.unique(arr)
+
+        assert isinstance(result, np.ndarray)
+        assert len(result) == 3
+
     def test_on_index_object(self):
 
         mindex = pd.MultiIndex.from_arrays([np.arange(5).repeat(5), np.tile(