diff --git a/docxtpl/inline_image.py b/docxtpl/inline_image.py
index f860749..5b164b1 100644
--- a/docxtpl/inline_image.py
+++ b/docxtpl/inline_image.py
@@ -4,7 +4,7 @@
 
 @author: Eric Lapouyade
 """
-from docx.oxml import OxmlElement, parse_xml
+from docx.oxml import OxmlElement
 from docx.oxml.ns import qn
 
 
@@ -19,11 +19,24 @@ class InlineImage(object):
     width = None
     height = None
     anchor = None
-
-    def __init__(self, tpl, image_descriptor, width=None, height=None, anchor=None):
+    title = None
+    descr = None
+
+    def __init__(
+        self,
+        tpl,
+        image_descriptor,
+        width=None,
+        height=None,
+        anchor=None,
+        title=None,
+        descr=None,
+    ):
         self.tpl, self.image_descriptor = tpl, image_descriptor
         self.width, self.height = width, height
         self.anchor = anchor
+        self.title = str(title) if title is not None else None
+        self.descr = str(descr) if descr is not None else None
 
     def _add_hyperlink(self, run, url, part):
         # Create a relationship for the hyperlink
@@ -49,19 +62,30 @@ def _add_hyperlink(self, run, url, part):
 
         return run
 
+    def _set_doc_properties(self, run):
+        docPr = run.xpath(".//wp:docPr")[0]
+        cNvPr = run.xpath(".//pic:cNvPr")[0]
+
+        for elt in (docPr, cNvPr):
+            if self.title is not None:
+                elt.set("title", self.title)
+            if self.descr is not None:
+                elt.set("descr", self.descr)
+
     def _insert_image(self):
         pic = self.tpl.current_rendering_part.new_pic_inline(
             self.image_descriptor,
             self.width,
             self.height,
-        ).xml
+        )
+        if self.title is not None or self.descr is not None:
+            self._set_doc_properties(pic)
         if self.anchor:
-            run = parse_xml(pic)
-            if run.xpath(".//a:blip"):
-                hyperlink = self._add_hyperlink(
-                    run, self.anchor, self.tpl.current_rendering_part
+            if pic.xpath(".//a:blip"):
+                pic = self._add_hyperlink(
+                    pic, self.anchor, self.tpl.current_rendering_part
                 )
-                pic = hyperlink.xml
+        pic = pic.xml
 
         return (
             "</w:t></w:r><w:r><w:drawing>%s</w:drawing></w:r><w:r>"
diff --git a/tests/inline_image.py b/tests/inline_image.py
index 48ed670..aa2b046 100644
--- a/tests/inline_image.py
+++ b/tests/inline_image.py
@@ -5,6 +5,9 @@
 @author: Eric Lapouyade
 """
 
+import zipfile
+from xml.etree import ElementTree
+
 from docxtpl import DocxTemplate, InlineImage
 
 # for height and width you have to use millimeters (Mm), inches or points(Pt) class :
@@ -14,7 +17,13 @@
 tpl = DocxTemplate("templates/inline_image_tpl.docx")
 
 context = {
-    "myimage": InlineImage(tpl, "templates/python_logo.png", width=Mm(20)),
+    "myimage": InlineImage(
+        tpl,
+        "templates/python_logo.png",
+        width=Mm(20),
+        title="python-logo-title",
+        descr="python-logo-description",
+    ),
     "myimageratio": InlineImage(
         tpl, "templates/python_jpeg.jpg", width=Mm(30), height=Mm(60)
     ),
@@ -49,3 +58,26 @@
 jinja_env = jinja2.Environment(autoescape=True)
 tpl.render(context, jinja_env)
 tpl.save("output/inline_image.docx")
+
+
+with zipfile.ZipFile("output/inline_image.docx") as docx_file:
+    document = ElementTree.fromstring(docx_file.read("word/document.xml"))
+
+namespaces = {
+    "pic": "http://schemas.openxmlformats.org/drawingml/2006/picture",
+    "wp": "http://schemas.openxmlformats.org/drawingml/2006/wordprocessingDrawing",
+}
+
+picture = document.find(".//pic:cNvPr[@name='python_logo.png']", namespaces)
+if picture is None:
+    raise ValueError("Missing python_logo.png image metadata")
+if picture.get("title") != "python-logo-title":
+    raise ValueError("Missing picture title metadata")
+if picture.get("descr") != "python-logo-description":
+    raise ValueError("Missing picture description metadata")
+
+doc_pr = document.find(".//wp:docPr[@title='python-logo-title']", namespaces)
+if doc_pr is None:
+    raise ValueError("Missing wp:docPr title metadata")
+if doc_pr.get("descr") != "python-logo-description":
+    raise ValueError("Missing wp:docPr description metadata")