diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py
index 19370f62b..b3935518c 100755
--- a/youtube_dl/YoutubeDL.py
+++ b/youtube_dl/YoutubeDL.py
@@ -864,6 +864,8 @@ class YoutubeDL(object):
         elif result_type == 'url':
             # We have to add extra_info to the results because it may be
             # contained in a playlist
+            if ie_result.get('channel_image'):
+                extra_info['channel_image'] = ie_result.get('channel_image')
             return self.extract_info(ie_result['url'],
                                      download,
                                      ie_key=ie_result.get('ie_key'),
@@ -998,6 +1000,7 @@ class YoutubeDL(object):
                     'extractor_key': ie_result['extractor_key'],
                 }
 
+                extra.update(extra_info)
                 reason = self._match_entry(entry, incomplete=True)
                 if reason is not None:
                     self.to_screen('[download] ' + reason)
diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index b35bf03aa..527cf1922 100644
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -2962,6 +2962,8 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
         if channel_page is False:
             channel_playlist_id = False
         else:
+            channel_image = self._html_search_meta(
+                'og:image', channel_page, 'channel image', default=None)
             channel_playlist_id = self._html_search_meta(
                 'channelId', channel_page, 'channel id', default=None)
             if not channel_playlist_id:
@@ -2974,8 +2976,10 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
                         channel_url, 'channel id', default=None)
         if channel_playlist_id and channel_playlist_id.startswith('UC'):
             playlist_id = 'UU' + channel_playlist_id[2:]
-            return self.url_result(
+            url_result = self.url_result(
                 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
+            url_result['channel_image'] = channel_image
+            return url_result
 
         channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
         autogenerated = re.search(r'''(?x)